Description of the embodiments
The technical scheme of the application is clearly and completely described below with reference to the examples and the drawings. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
Lidar system and spatial suspension interference problem
Lidar systems are used to measure the environment surrounding them and have been widely used for autonomous navigation, spatial mapping, and the like. Lidar systems emit pulses of light that are reflected by objects in the environment, and the distance between the object and the system is determined by measuring the time of flight of the light pulses from transmission to reception.
As shown in fig. 1a, a laser radar according to an embodiment of the present application includes a transmitter 101, a collector 102, and a control and readout circuit 103 connected to the transmitter and the collector. Transmitter 101 is configured to transmit optical signal 110 toward target 112, and echo signal 111 reflected by target 112 is collected by collector 102 and then output an electrical signal. A control and readout circuit 103 is connected synchronously with the transmitter 101 and the collector 102 for controlling the transmitter 101 to activate the collector 102 to receive the echo signals while transmitting the light pulses and to perform a data readout function to process the echo signals to output distance data of the target. In some embodiments, lidar 10 also includes data processing circuitry 104 for receiving range data to generate point cloud data for a detection space.
As shown in fig. 1b, a laser radar with a scanning mirror according to another embodiment of the present application, the laser radar 10 includes a transmitter 101, a collector 102, a scanning unit 107, a control and readout circuit, and a data processing circuit (not shown). Wherein the scanning unit 107 may comprise at least one scanning element, which may be OPA, liquid crystal, mirror, turning mirror, MEMS mirror, galvanometer, etc. For example, as shown in fig. 1b, the scanning unit comprises a turning mirror 106, and during the detection process, the turning mirror 106 rotates at a very fast speed, so as to change the propagation path of the emitted light signal 110 to be projected onto different areas of the detection space, thereby realizing the scanning of the target 112; the echo signal 111 reflected by the target 112 is transmitted to the collector 102 after being reflected by the turning mirror 106, the electric signal is output after being collected by the collector, the electric signal is read by the control and readout circuit to obtain target distance data, and the data processing circuit analyzes and processes the target distance data to generate point cloud data of a detection space. In some embodiments, to expand the optical path, the optical system further includes a mirror 105, and the optical signal 110 emitted by the emitter 101 is transmitted to the mirror 105 and then reflected to the turning mirror 106, and reflected to the target 112 through the turning mirror; likewise, the reflected optical signal 111 is coupled into the detector 102 via the turning mirror 106 and the mirror 105.
The details of each of these parts will be described below in connection with the schematic diagrams shown in fig. 1a and 1 b.
The emitter 101 includes a light source, an emission driving circuit, and an emission optical element. The emission driving circuit is used for driving the light source to emit light signals, and the light signals are modulated by the emission optical element and then emitted to the detection space. The light source may be a Vertical Cavity Surface Emitting Laser (VCSEL), an Edge Emitting Laser (EEL), a Light Emitting Diode (LED), or other type of laser. The emission optical element includes one or more of a lens, a microlens array, a diffraction optical element, a diffusion sheet, a liquid crystal, and the like. In the detection space, the light beam emitted by the light source is a light beam with a certain divergence angle, the emitted light beam is modulated by the emission optical element, one or more illumination areas (such as one or more light spots) are formed after the light beam reaches the detection space, the area of the light spots is larger and larger along with the continuous increase of the distance, and the covered target area is also gradually enlarged.
The collector 102 includes a photosensitive element, a receiving optical element, and the like, where the receiving optical element images the light spot beam reflected by the target onto the photosensitive element, and the photosensitive element is configured to collect the reflected light beam, convert the light signal into an electrical signal, and output the electrical signal, where information (such as echo intensity information, time information, and the like) of the electrical signal reflects the light beam, and may be further used to calculate information such as photon flight time, target distance, target reflectivity, and the like. The photosensitive element generally comprises one or more pixel units, the pixel can be one of APD, SPAD, siPM devices for collecting optical signals, the receiving optical element comprises an optical filter and an imaging lens, the optical filter is configured to be matched with the wavelength of the emitted optical signals so as to filter out other interference optical signals, and the imaging lens is used for receiving the optical signals and converging the optical signals to the photosensitive element. In one embodiment, the photosensitive element is a SPAD, and multiple SPADs may be optionally combined to collect optical signals for increasing the received energy due to the smaller SPAD size, e.g., a photosensitive element includes 4*4 SPADs for collecting optical signals of a single spot.
The control and readout circuit 103 is used to control the synchronous activation of the emitter 101 and collector 102, and to receive and process the electrical signals output by the photosensitive elements to calculate the photon flight time, and thereby obtain the target distance. In one embodiment, the photosensitive element is a SPAD capable of detecting weak optical signals; the control and readout circuit 103 includes a TDC circuit, a histogram circuit, and a digital signal processing circuit. The TDC circuit receives the laser emission signal and records the time of flight of photons from emission to collection and generates a time signal (e.g., a time code) which is input to the histogram circuit and stored in a corresponding storage unit (time bin). After repeated measurements, histogram data containing pulse waveforms and photon times of flight are obtained based on time-dependent single photon counting (TCSPC). The digital signal processing circuit is used for processing the histogram data to calculate the echo intensity and the photon flight time, and can further acquire the target distance according to the photon flight time. In some embodiments, the control and readout circuitry 103 may also be integrated into the collector as part of which the control and readout circuitry 103 is included for ease of description in the embodiments described below.
In the laser radar measurement process, a light spot beam emitted towards a detection space is projected to a fixed position in the space, reflected by a target and imaged to a corresponding photosensitive element, so as to form a corresponding relationship, which is commonly referred to as "forming a detection channel" in the art. The control and readout circuit 103 receives the electrical signal output from the photosensitive element and processes it to obtain the distance of the target in the detection channel.
In some embodiments, to meet a larger field angle and a faster scanning to improve detection resolution and frame rate, the lidar may be configured to use multi-channel measurement, that is, the emitter 101 may be configured to include a light source array including a plurality of light sources and the collector 102 may be configured to include a light sensing element array including a plurality of light sensing elements, the light sensing elements are set as minimum detection units of the collector, the number of light sensing elements is adapted to the number of emitted light spot beams, the light source array and the light sensing element array are correspondingly arranged to form a plurality of detection channels, and each detection channel includes at least one light source and at least one light sensing element, where the light source array and the light sensing element array may be one-dimensional arrays or two-dimensional arrays. The control and readout circuitry 103 is configured as an array of circuits comprising a plurality of sub-circuits, each for processing the electrical signals output by the photosensitive elements in each detection channel to obtain the distance of the target in that channel. Thus, the control and readout circuit 103 can pre-identify the light source and the photosensitive element corresponding to each detection channel, and can reduce the system power consumption according to the operation mode of regulating and controlling the light source and the photosensitive element.
In some embodiments, without controlling the entire light source array to emit the spot beam toward the detection space at the same time, only a subset of the light source array and a corresponding subset of the light sensing element array may be selected to be activated at a time to obtain a distance of at least one target, so as to reduce power consumption and complexity of the circuit; then, activating different subsets of the light source array at different times, and simultaneously acquiring distances of different targets detected by the different light-sensing element array subsets; finally, all the light sources in the light source array can be activated in a scanning mode to finish scanning measurement of the detection space. The subset may comprise at least one light source and at least one light sensitive element forming at least one detection channel in an operative state. In one embodiment, the array of light sources may be configured in a row-by-row, column-by-column manner, e.g., the array of light sources may be activated one column at a time and sequentially emit light in a left-to-right fashion, while the array of light sensing elements is also configured to activate acquisition echo signals in a corresponding order. In another embodiment, the light source array may be configured to be divided into a plurality of light source groups, and as such, the light-sensing element array is configured to be divided into a plurality of light-sensing groups, with each simultaneously activated light source being located within a different light source group, and with simultaneously activated light-sensing elements being located within different light-sensing groups.
When the detection of the detection space is completed, the calculated distance data is input into the data processing circuit 104, and the data processing circuit 104 is used for generating 3D point cloud data of the detection space after performing processes such as filtering and analysis on the data. The data processing circuit 104 may be a separate dedicated circuit, such as a dedicated SOC chip, FPGA chip, ASIC chip, or the like, or may include a general-purpose processor. In one embodiment, the data processing circuit 104 may be integrated with the control and readout circuit 103 and integrated into the photosensitive element array as part of the collector, for example, by configuring the photosensitive element array as a stacked structure, with the photosensitive element on top, and the corresponding control and readout circuits and data processing circuits on the bottom. In another embodiment, the data processing circuit 104 may also be part of a processor in the terminal when the lidar is integrated into the smart terminal.
In some embodiments, the control and readout circuit 103 may further output an echo intensity, and the data processing circuit 104 further includes a reflectivity calculation unit for calculating a reflectivity of the target based on the echo intensity and the distance of the target. Specifically, the target reflectivity, the echo intensity and the distance satisfy the following relation:
Where I is the echo intensity, d is the distance data, and k is set according to system parameters, and is mainly related to the emitted light signal intensity, the outgoing and receiving efficiency of the lens, the diameter of the light spot reaching the target, the light transmission efficiency η transmitted from the receiving element to the sensing area, and the photoelectric conversion efficiency PDE of the photosensitive element array.
In some embodiments, when the digital signal processing circuit processes the histogram data to perform the echo intensity and distance data calculation, effective echo intensity and distance data may not be obtained due to some special cases, such as no pulse being identified in the histogram, or the number of pulses not meeting the calculation condition being identified, etc., since the number of bits of the output data of the control and readout circuit 103 is fixed, the digital signal processing circuit may assign the echo intensity and distance data to predetermined values (such as the predetermined value being set to be greater than the maximum value of the system detection capability). Accordingly, the data output from the control and readout circuit 103 may also be configured to include a variety of information so that the resolution can be identified when the data processing circuit 104 reads the data information. In one embodiment, the data output by the configuration control and readout circuit 103 includes common information including probe channel numbers, invalid information, state of solution, etc., and solution information including distance data, echo intensity data, etc.; the data processing circuit 104 reads and processes the data output by the control and readout circuit 103 based on a predefined protocol to generate 3D point cloud data of the detection space. Wherein, there is invalid information for determining whether the solution distance information is valid and invalid, for example, the predefined invalid condition includes that the ambient light is overexposed, the pulse number exceeds a threshold value, the signal to noise ratio is too low, and the like, and the data processing circuit can determine whether the data is valid and invalid by reading the invalid information.
In some embodiments, the data processing circuit 104 further includes a storage unit for storing a pre-calibrated reflectivity map to calculate the reflectivity of each target from the reflectivity map. In one embodiment, diffuse reflection plates with different reflectivities are placed at a preset distance dc, and the laser radar measures each diffuse reflection plate to obtain a series of corresponding echo intensities I, for example, the diffuse reflection plates are placed at 5m, and the diffuse reflection plates with the reflectivities of 0.1, 0.2, 0.5, 0.9, 1, 10 and the like are selected, and the echo intensities at each reflectivity are detected to obtain a data setI=1, 2 … n, based on which a reflectance mapping table is formedAnd stored. In an actual measurement process, a set of measured echo intensities and distances are obtained, e.g.Mapping the group of data to a preset distance to calculate the equivalent echo intensityAccording to the equivalent echo intensityThe closest I value is searched in the reflectivity mapping table, and the corresponding reflectivity rho is further calculated based on linear interpolation.
The laser radar can be used on terminals such as autonomous or semi-autonomous road vehicles, aircrafts, ships and the like to obtain surrounding environment information, and further realize functions of navigation, obstacle avoidance, identification and the like, but in actual use, many complex situations can be faced, such as when severe weather such as rain, snow, haze and the like is encountered, a large number of raindrops, snowflakes or haze particles exist in the air in a detection space (for convenience of analysis, the difference of the optical flight speed and the movement speed of the particles such as rain, fog, snow and the like are considered at the same time, the particles such as rain, fog, snow and the like are regarded as space suspension). In order to realize long-distance detection capability, such as 200m, the laser radar generally regulates and controls a smaller light beam emergence angle, so that more concentrated light spot energy is ensured, and the diameter of a light spot projected by a light beam gradually increases from a few millimeters to tens of millimeters along with the distance. In the detection space, the minimum size of a conventional object is generally about 30-50mm, the size of rain and snow particles is about 3-6mm, the maximum size of the rain and snow particles is generally not more than 6mm, and the rain and snow particles are transparent and sparse in space (with space sparsity). Therefore, if the light beam irradiates the rain and snow particles at a far distance, the formed light spot size is far larger than the particle size, the reflected echo energy is lower and can not be collected by the photosensitive element, or can be collected by the photosensitive element, the echo intensity is very weak compared with the echo signal of the conventional object, and the target reflectivity can be calculated to distinguish the conventional object. However, at a relatively short distance, for example, within <10m, the spot size of the light beam irradiated on the rain and snow particles is similar to the particle size, so that a relatively strong reflected signal is formed. Particularly, for the case of snow, the snow is crystalline and has higher reflectivity, a stronger echo signal can be formed, and a dispersion state can be presented on the photosensitive element, so that the imaging characteristic of the echo signal is similar to that of a conventional object and cannot be distinguished. The haze belongs to small particle suspended matters in an aggregation state, light spots at close range cover a plurality of haze particles, echo signal characteristics similar to those of conventional objects appear, and false recognition can be generated. The special conditions appear, so that a large number of noise points exist in the point cloud data output by the laser radar, namely, the situation that the space suspended matters are mistakenly identified as targets occurs, the ranging capability of the laser radar is affected, and the measurement reliability of the laser radar is reduced.
In addition, the laser radar needs to emit high-power beam signals in order to expand the detection range, and the dynamic range of the collector is limited, so that signal saturation is easy to occur during short-distance detection, deviation occurs in echo intensity measurement, and spatial suspended matters cannot be accurately identified.
The existing solutions mainly comprise point cloud filtering algorithms or system level improvements to reduce the influence of spatial suspensions on measurement data. The point cloud filtering algorithm is characterized in that the distribution of the point cloud of a normal target object is continuous, and the spatial suspended matters such as rain, snow, haze and the like have the spatial discreteness due to small volume and discrete distribution, so that noise points can be judged according to the distribution of the point cloud. However, in practical applications, it may be difficult to distinguish discontinuous small objects from spatial suspensions (such as haze, etc.), and misidentification may still occur easily. In the improvement of the system level, the method mainly comprises the steps of firstly obtaining complete detection data under the whole visual field, identifying the position of a suspected noise point, and then carrying out secondary measurement on the position to distinguish a space suspension object from a target. Under the condition, the method can effectively solve the problem of middle-distance or long-distance noise points, but the laser radar cannot output effective detection signals due to overexposure of signals in short distance, and the method cannot distinguish the noise points from targets, and in addition, the measurement efficiency is reduced due to secondary measurement.
(II) double laser radar model and principle of identifying space suspended matters
Aiming at the problems in the prior art, in the scheme of the application, a double-laser radar model is provided, which comprises an auxiliary radar for emitting a large-spot light beam and a main radar for emitting a small-spot light beam, and the double radars are matched to solve the problem that a space suspended object and a conventional object are difficult to partition. It should be noted that, the dual radar, the primary radar, and the secondary radar mentioned herein are only convenient for description and are not limited to the actual product, the system architecture, the structural configuration, the module composition, and the like in the technical solution, and may be two working configurations of one radar or two units of one radar or two independent lidars. Hereinafter, the scheme of the present application will be described in detail.
The double-laser radar model comprises a main radar, an auxiliary radar and a data processing circuit, wherein the main radar and the auxiliary radar emit light beams to the same detection target, and the main radar emits small light spot light beams towards the target and collects the small light spot light beams reflected by the target to generate a first distance; the secondary radar emits a large spot beam toward the target and collects the large spot beam reflected back by the target to generate a second distance and a second echo intensity. The data processing circuit generates a second target reflectivity according to the second distance and the second echo intensity, judges whether the target is a space suspension or not by utilizing the first distance, the second distance and the second target reflectivity, and optimizes the first distance according to a judging result. Typically, when the object is determined to be a spatial suspension, the first distance of the object is removed, thereby obtaining distance data of the object.
The large-spot light beam and the small-spot light beam are light beams emitted by the auxiliary radar, namely, the light beam emitted by the auxiliary radar irradiates the target to form a light spot with a larger size than the light spot formed by the light beam emitted by the main radar irradiating the target, so that the light beams emitted by the auxiliary radar are called large-spot light beams, and the light beams emitted by the main radar are called small-spot light beams for distinguishing. The small-spot beam and the large-spot beam are the opposite concepts, and are not limited thereto. The number of beams is not limited, and may refer to one beam or a plurality of beams, such as a large spot beam and a small spot beam each including a plurality of sub-beams arranged in an array in some embodiments.
Based on a double-laser radar model, the application provides a distance detection method, which specifically comprises the following steps:
s1: emitting a small light spot beam and a large light spot beam to the same detection target;
S2: receiving the small light spot beam reflected back by the target and generating a first distance, and receiving the large light spot beam reflected back by the target and generating a second distance and a second echo intensity;
s3: generating a second target reflectivity according to the second distance and the second echo intensity;
S4: judging whether the target is a space suspension or not by utilizing the first distance, the second distance and the second reflectivity;
s5: and optimizing the first distance according to the judging result to obtain the distance information of the target.
In some embodiments, the target is a regular object when the second target reflectivity is greater than the second reflectivity cut-off threshold and the distance deviation is not greater than the distance threshold by comparing the target reflectivity to the second reflectivity cut-off threshold and comparing the distance deviation of the first distance and the second distance to the distance threshold. In other cases, the target is a spatial suspension. And when the target is a space suspension, eliminating the first distance corresponding to the target. Wherein the second reflectivity cut-off threshold is set according to parameters of the secondary radar.
In some embodiments, if the first distance corresponding to the small spot beam is valid and the second distance corresponding to the large spot beam and the second echo intensity are invalid, then it is determined that the target corresponding to the small spot beam is a spatial suspension. And when the target is a space suspension, eliminating the first distance corresponding to the target. Typically, the second distance and second echo intensity inefficiency is that the signal-to-noise ratio is too low to obtain the distance and echo intensity for histogram data processing.
In some embodiments, the size of the emitted large spot beam is 4-1000 times the size of the small spot beam. Since the size of the large spot beam is larger than that of the small spot beam, when the multi-channel detection is adopted, the main radar is configured to emit a plurality of small spot beams toward the detection space, and the auxiliary radar is configured to emit at least one large spot beam toward the detection space, each large spot beam covering the plurality of small spot beams.
In the scheme of the application, the auxiliary radar and the main radar are matched to identify the space suspended matters and the conventional objects, and compared with the auxiliary radar, the main radar can quantify the difference of the echo intensities reflected by the conventional objects and the space suspended matters at the same distance, so that the conventional objects and the space suspended matters can be distinguished. Aiming at the condition that the main radar detects overexposure in a short distance, the auxiliary radar can also ensure that overexposure cannot occur when the target is a space suspension, and the reflectivity of the target can be obtained.
The above-described dual lidar model and range detection method will be exemplarily described with reference to fig. 2.
As shown in fig. 2, a schematic diagram of a dual lidar model according to an embodiment of the present application is described for convenience in describing a single probe channel, and may include multiple probe channels in practical applications. The dual lidar model comprises a primary radar 21, a secondary radar 22 and a data processing circuit 23, which cooperate to detect a target space, and will be described below by taking the case of snowflake targets appearing in the detection space as an example, and in addition for simplicity, and only the light source and the photosensitive element are shown in the drawings, while other devices such as optical elements are omitted, specific constituent parts can be referred to fig. 1, and the composition of each part of the lidar shown in fig. 1 is also included in the present embodiment.
The main radar 21 is a range radar, and includes a first light source 211 and a first photosensitive element 212. In order to satisfy the long-distance detection, for example, the detection distance exceeding 200m, the emission angle of the light beam emitted by the first light source 211 needs to be not more than 0.3 degrees, for example, 0.2 degrees, the emitted light beam is transmitted to the detection space through the emission optical element (such as a lens) to form a light spot, the size of the light spot gradually increases along with the increase of the distance, for example, about 14mm at 0.5m, and the size of the light spot at 10m is about 37mm. If snowing weather occurs during detection, snow may appear in the detection channel, as shown in fig. 2, if the snow appears in a short distance, for example, 0.5m, the size of the snow is about 6mm, the size of the light spot is similar to that of the snow, the light spot can basically cover the snow so that laser energy irradiates the snow, the snow is crystalline and has higher reflectivity, a stronger echo signal can be formed by the reflected laser signal, and the reflected light signal presents a dispersion state when entering the first photosensitive element 212, so that the echo intensity is similar to that of a conventional object, and whether the target is a space suspension cannot be distinguished. If the snow appears at 10m, the size of the light spot is larger than that of the snow, and only part of the laser energy irradiates the snow, most of the energy can penetrate through the snow to irradiate a rear conventional object. Because the size of the conventional object is far larger than that of the snowflake, the echo intensity of the snowflake reflected back at the distance is far smaller than that of the conventional object, and the space suspension object and the conventional object can be distinguished, typically, the space suspension object or the conventional object can be judged by calculating the reflectivity of the target. In addition, since the detection distance of the main radar 21 is far, the laser power of the beam emitted by the first light source 211 needs to be configured to be increased, and overexposure is easy to occur in a short-distance range, at this time, the echo intensities of the targets detected by the main radar 21 are similar, and there is no method for identifying noise.
Based on this, the application configures the auxiliary radar to work cooperatively. The auxiliary radar 22 includes a second light source 221 and a second photosensitive element 222, and the second light source 221 is configured to emit light beams toward the same detection target as the first light source 211, and the exit angle of the light beams emitted by the second light source 221 is configuredIs larger than the emergent angle of the light beam emitted by the first light source 211For example 2 deg., such that the spot size of the light beam emitted by the second light source 221 is larger than the spot size of the light beam emitted by the first light source 211, i.e. the second light source 221 emits a large spot light beam and the first light source 211 emits a small spot light beam. In addition, the photosensitive elements 212 and 222 are respectively adapted to the sizes of the spot beams emitted by the light sources 211 and 221, and the photosensitive area of the second photosensitive element 222 is larger than that of the first photosensitive element 212. The second light source 221 emits a light beam toward the same detection target, and the size of a large spot light beam formed when the light beam irradiates the snowflake at 0.5m is about 22mm, which is significantly larger than the spot size of the snowflake. Therefore, only a part of the laser energy is reflected to form an echo signal to be incident on the second photosensitive element 222, and the echo intensity output by the auxiliary radar 22 is greatly different from that of the conventional object, so that the spatial suspended object and the conventional object can be distinguished according to the data output by the auxiliary radar.
When the large-spot light beam and the small-spot light beam are configured to irradiate the same target, if the target is a conventional object, the reflected echo intensity is approximate because the conventional object has a large size; if the object is a space suspension, only a very small part of the light spot energy in the large light spot beam irradiates the particles, and the echo intensities generated by the two light spot energy are different. Therefore, the reflectivity of the target can be accurately calculated by utilizing the data output by the auxiliary radar, so that whether the target is a space suspension or a conventional object can be judged according to the reflectivity of the target and the distance data output by the main radar and the auxiliary radar, and the distance data output by the main radar is optimized according to the judgment result to obtain the distance data of the target. Further, the distance targeted to the spatial suspension is eliminated.
Specifically, the data processing circuit receives a first distance output by the main radar, a second distance output by the auxiliary radar and a second echo intensity, and calculates a second target reflectivity according to the second distance and the second echo intensity; and comparing the second target reflectivity to a reflectivity cutoff threshold and comparing a distance deviation of the first distance and the second distance to a distance threshold. Only when the distance deviation between the first distance and the second distance is not greater than the distance threshold and the reflectivity of the second target is greater than the reflectivity cut-off threshold, the target corresponding to the first distance can be determined to be a conventional object, and under other conditions, the targets corresponding to the first distance are all space suspended matters, and further the first distance judged to be the space suspended matters can be removed. I.e. the object is considered a regular object when the following relation needs to be satisfied:
Wherein d1 is the first distance of the primary radar output and d2 is the second distance of the secondary radar output; Is a distance threshold; Is the second reflectivity cut-off threshold.
The laser radar has measurement deviation when measuring the target distance, and is mainly determined by factors such as system parameters, random errors and the like. Considering that the parameters of the main radar system and the auxiliary radar system are not identical, the distance data respectively output when the main radar and the auxiliary radar measure the same target are not necessarily identical, namely the following relation is satisfied:
Wherein,AndAnd the system measurement deviation values corresponding to the main radar and the auxiliary radar respectively, wherein D is the actual distance of the target. However, since the two data belong to the same object, if the object is a conventional object, the deviation of the two distance data should satisfy a certain threshold value, i.eThus, in some embodiments, the first and second processing elements,Can be set as 。
In some embodiments of the present invention, in some embodiments,The reflectivity cut-off threshold value set based on the auxiliary radar parameters is set through pre-calibration to distinguish the conventional object from the spatial suspension.
In some embodiments, as shown in fig. 2, if rainy weather occurs at the time of detection, raindrops appear at 0.5m in the detection channel, if the size of the raindrops at this time is small, for example, only 1mm. The small spot size of the beam emitted by the main radar 21 at 0.5m is about 14mm, and the spot irradiates the raindrop to form an effective echo signal and outputs a first distance. The large spot size of the beam emitted by the auxiliary radar 22 irradiated to the raindrop is about 22mm, only a very small part of laser energy is reflected, the reflected laser energy may be submerged in the ambient light, and the signal-to-noise ratio is too low to enable the auxiliary radar to detect the effective second distance and the second echo intensity, so that the data information output by the auxiliary radar is invalid. When the data processing circuit receives and processes the data information output by the main radar and the auxiliary radar, the data information is read to determine that the first distance is effective, the second distance is ineffective, and the second echo intensity is ineffective (the signal to noise ratio is too low), at the moment, the second target resolution can be directly determined instead of generating the second target resolution according to the second distance and the second echo intensity, the target corresponding to the small light spot beam is a space suspension, and the first distance is removed. It will be appreciated that by calculating the second target reflectivity in combination with the first and second distances, it can also be determined whether the target is spatially suspended, but this will undoubtedly increase power consumption.
In some embodiments, since the situation that the spatial suspension and the conventional object cannot be resolved in the main radar is mainly in a close range, and the ranging range of the auxiliary radar is smaller than that of the main radar in consideration of the difference of the system saturation cut-off distances corresponding to the difference of the ranging capability of the systems. For example, setting the range of the main radar to be 200m, and determining the saturation cut-off distance of the system to be 20m according to the limit ranging condition; if the auxiliary radar measurement range is set to be 10m, the system saturation cut-off distance is set to be 1m under the same limit ranging condition; the corresponding signal saturation levels are all 1. For the embodiment shown in fig. 2, when snowflake at 0.5m is detected, the reflectivity of the snowflake is set to 1, the signal saturation of the first photosensitive element 222 in the main radar may be about 289, the signal saturation far exceeds the set signal saturation, and the first echo intensity output in the main radar is inaccurate, so that the target reflectivity cannot be calculated. The saturation level of the second light sensing element 321 in the auxiliary radar is about 0.288, which is far smaller than the saturation level of the signal set by the system, and for the echo signal reflected by the same target, the light sensing element of the auxiliary radar is in an unsaturated state, and the detected second echo intensity is an accurate value, so that the reflectivity of the target can be accurately calculated. Therefore, the second target reflectivity can be calculated according to the second echo intensity and the second distance data output by the auxiliary radar, and the distance data output by the main radar is analyzed according to the second target reflectivity and the distance data of the target points detected by the main radar and the auxiliary radar, so that the first distance corresponding to the space suspension is deleted, and the detection accuracy of the system is improved.
(III) anti-dry anti-laser radar system
Based on the foregoing principle, the present application further proposes a laser radar system, which includes a main radar module, an auxiliary radar module, and a data processing circuit, where the data processing circuit is configured to perform distance detection on N sub-areas in a target space, and process and analyze data output by the main radar module and the auxiliary radar module to generate 3D point cloud data with higher accuracy.
Specifically, the main radar module includes a main transmitter including a first light source array and a first transmitting optical element for transmitting a plurality of small spot light beams toward each sub-area of the detection space, and a main collector including a first light sensing element array, a first receiving optical element, and a first control and readout circuit for receiving an echo signal reflected by the target and outputting a first distance, and also outputting a first echo intensity of the echo signal. Similarly, the auxiliary radar module comprises an auxiliary emitter and an auxiliary collector, wherein the auxiliary emitter comprises a second light source array and a second emission optical element and is used for emitting a large-spot light beam towards each sub-area of the detection space, and the auxiliary collector comprises a second photosensitive element array, a second receiving optical element and a second control and readout circuit and is used for receiving an echo signal reflected by a target and outputting a second distance and also outputting a second echo intensity of the echo signal. Further, the data processing circuit receives second distance calculation and second echo intensity output by the auxiliary radar to calculate a second target reflectivity corresponding to each large-spot light beam, judges whether a target corresponding to each small-spot light beam is a space suspension or not according to the second target reflectivity, the first distance and the second distance, and optimizes the first distance according to a judgment result to obtain distance information of the target.
In some embodiments, the data processing circuit compares the second target reflectivity to a second reflectivity cut-off threshold, and compares a distance deviation of the first distance and the second distance to a distance threshold, the target being a regular object when the second target reflectivity is greater than the second reflectivity cut-off threshold and the distance deviation is not greater than the distance threshold. In other cases, the target is a spatial suspension. Further, when the target is a spatial suspension, the data processing circuit rejects the first distance of the target. The second reflectivity cut-off threshold is a reflectivity cut-off threshold of the auxiliary radar module set through pre-calibration.
In some embodiments, the setting of the spot size, e.g. the light emission aperture, may be achieved by configuring parameters of the light source arrays, i.e. the light emission aperture of a first light source in a first light source array is larger than the light emission aperture of a second light source in a second light source array. In another embodiment, the second light source array may be located out of focus from the second emission lens, i.e. the first light source array is located at a distance from the first emission lens equal to the focal length of the first emission lens, and the second light source array is located at a distance from the second emission lens greater than the focal length of the second emission lens. Through defocusing setting, the light beam emitted by the second light source diverges and projects to the detection space through the second emission lens to form a large-spot light beam. In some embodiments, the size of the large spot beam is about 4-1000 times the size of the small spot beam.
In some embodiments, the detection range of the secondary radar module is less than the detection range of the primary radar module.
Because the laser radar system is configured to adopt a multi-channel detection mode, and the large-spot light beam is larger than the small-spot light beam, in the detection process, the field of view irradiated by each large-spot light beam can correspondingly cover the fields of view irradiated by a plurality of small-spot light beams; on the other hand, based on the multiple scanning modes during the multi-channel detection, the detection time sequences of the main radar module and the auxiliary radar module are staggered. Therefore, alignment calibration is required to be carried out on the position relation of the emitted light spots of the main radar module and the auxiliary radar module irradiated to the detection space so as to determine the corresponding relation of the light spots with the size, and subsequent data processing is convenient.
Fig. 3 is a schematic diagram of scanning of a lidar system according to an embodiment of the present application. The lidar system detects a detection space comprising N sub-areas, such as the 4 sub-areas 31, 32, 33, 34 in fig. 3. The main radar module is configured to emit m small spot beams toward each sub-area, for example, the main radar module emits 3*3 small spot beams toward each sub-area, and the auxiliary radar module emits one large spot beam toward each sub-area for detection, i.e., each 9 small spot beams corresponds to one large spot beam.
According to the division of the subareas and the spatial position relationship between the large light spot beam and the small light spot beam projected into each subarea, the corresponding relationship between the large light spot beam and the small light spot beam can be marked, for example, in fig. 3, each large light spot beam corresponds to 9 small light spot beams. The data processing circuit needs to judge the first distance corresponding to each small light spot beam, namely, calculates a second target reflectivity according to the second distance corresponding to the large light spot beam in each sub-area and the second echo intensity, judges whether the targets detected by the 9 small light spot beams corresponding to each large light spot beam are space suspended matters or not by utilizing the first distance, the second distance and the second target reflectivity, and eliminates the first distance if the targets are the space suspended matters, and optimizes the first distance in each sub-area to obtain the distance information of the targets in the detection space.
It will be appreciated that the embodiment shown in fig. 3 is only schematically illustrated, and in other embodiments, the size of each sub-area may or may not be identical, and the size of the large spot beam corresponding to the different sub-areas may or may not be identical, and the size of the small spot beam of the different sub-areas may or may not be identical. In addition, one large-spot light beam in different subareas can correspond to different small-spot light beams, so that the method is not limited in any way, and can be adjusted based on actual needs. The data processing circuit can judge each large light spot beam and a plurality of corresponding small light spot beams one by one only by pre-calibrating the corresponding relation between the large light spot beam and the small light spot beam.
The conventional detection mode can control the main radar module to emit all small light spot beams to complete detection, then control the auxiliary radar module to emit all large light spot beams to complete detection, and the corresponding relation between the first distance and the second distance can be determined according to the division of the subareas and the spatial position relation of the pre-calibrated emitted light spots in the output data so as to judge whether each first distance is a conventional object or a spatial suspension. However, in order to improve the measurement accuracy, the ranging of the main radar module is often detected by adopting a scanning measurement mode, that is, a predetermined number of first light sources are selected to simultaneously activate to emit a plurality of small light spot beams to at least one sub-area, so that the auxiliary radar needs to be regulated and controlled to emit a large light spot beam according to the scanning mode of the main radar module and the position relation of the light spot projected to the detection space.
In some embodiments, the measurement may be performed by selecting a row-by-row/column-by-column scanning manner, and as shown in fig. 3, it is assumed that the main radar emits only one column of small spot light beams at a time, for example, the first emitted column spot a1 is projected into the sub-areas 31 and 33 to form the small spot light beams, so as to output measurement data once, and sequentially turn on the column spots a2 and a3 … until the last emitted column spot a6, that is, one frame of image data of the detection space is output. According to the corresponding relation of the pre-calibrated light spots, when the array light spot a1 is emitted, the first array light spot b1 emitted by the auxiliary radar module is required to be regulated and controlled to be projected into the subareas 31 and 33 to form a large light spot beam so as to output primary measurement data; when the column light spot a2 emits, the auxiliary radar data of the factor areas 31 and 33 are measured, the auxiliary radar module can not be regulated and controlled to emit light, or the first column light spot b2 can be regulated and controlled to emit light again, and the detection data can be updated. Likewise, the detection of the sub-regions 32, 34 may be performed in the same scanning manner. In other embodiments, the main radar module may also select an alternate or column lighting manner, for example, the first measurement may be performed by projecting the column light spots a1, a3, a5 into the sub-areas 31, 32, 33, 34 to form small light spot beams, and the auxiliary radar module may be controlled to project the column light spots b1, b2 into the sub-areas 31, 32, 33, 34 to form large light spot beams according to the light spot correspondence; the second measurement emits the column light spots a2, a3 and a6, and the auxiliary radar module can measure the updated data again or does not update.
In still other embodiments, to address the crosstalk between adjacent spots, the first array of light sources may be configured to be divided into multiple groups of light sources, with each simultaneously activated light source being located within a different group of light sources to emit multiple small spot beams toward different sub-areas. In one embodiment, the grouping of the first light source array may be configured according to the arrangement of the sub-areas, such as shown in fig. 3, where the light source array is divided into 4 groups according to four sub-areas 31, 32, 33, 34, and each sub-area corresponds to 9 small light spot beams as one group, and each group is only controlled to emit one light spot at a time, for example, the first light emission is performed so that the first light spot in each sub-area is projected, that is, the light spots a11, a41, a14, a44 are projected, and then projected one by one along a certain sequence to complete the detection. Correspondingly, when the light spots a11, a41, a14 and a44 are projected, the light spots b11, b21, b12 and b22 are projected towards the four sub-areas 31, 32, 33 and 34 by correspondingly regulating and controlling the auxiliary radars. When the secondary light emission is regulated, the corresponding auxiliary radar module can measure the updated data again or not. In another embodiment, the grouping of the first light source arrays may not be set according to the sub-areas, and only the space where the light beams of the small light spots are emitted may be considered, that is, the space where the light beams of the small light spots projected by the plurality of first light sources for controlling light emission are arranged at a certain interval when the light beams of the small light spots are irradiated to the detection space, so as to solve the crosstalk situation of adjacent light spots, for example, the adjacent light sources in the first light source arrays are configured as a group, and the first light sources for controlling light emission simultaneously are located in different light source groups so as to emit the light beams of the small light spots to different sub-areas. For example, the column light spots a1 and a2 are in one group, the column light spots a3 and a4 are in one group, and the column light spots a5 and a6 are in one group; modulating the first spot projection within each group, i.e. spots a11, a31, a51, to sub-areas 31 and 32, and modulating the emission of spots b11, b 21; the second light spot projection in each group is regulated, namely, light spots a21, a41 and a61 are projected to the subareas 31 and 32, and the corresponding auxiliary radar module can measure updated data again or can not update the updated data, and the space scanning is completed by sequentially projecting.
According to the foregoing embodiments, the activation manner and the corresponding scanning manner of the first light source array in the main radar do not need to be specifically configured according to the corresponding relationship between the sub-area and the large and small light spot beams, and are mainly set according to the arrangement manner, the light spot projection position, and the like of the first light source array in the main radar. According to the position relation of the irradiation of the large and small light spots in the detection space, no matter what scanning mode is adopted for regulating and controlling the main radar, after the main radar module finishes one-time measurement, namely, the auxiliary radar module is regulated and controlled to emit corresponding large light spot light beams according to the space position relation to finish one-time measurement, after the main radar module performs the second measurement, if the auxiliary radar module belongs to repeated measurement, whether data are updated or not can be considered according to the situation.
In the data transmission process, the main radar module and the auxiliary radar module can respectively splice the data to form a frame of finished image data to be output to the data processing circuit after the scanning measurement detection of the detection space is finished, or can also select to transmit the data once after the measurement is finished, and the measurement data of the main radar module and the auxiliary radar module can be synchronized by adopting a time stamp no matter how the data are transmitted, so as to determine the corresponding relation between the data. Therefore, the scanning method is not limited to the scheme of the application in the measuring process.
Based on the spatial position relation of light spot irradiation, the corresponding photosensitive element arrays in the main radar module and the auxiliary radar module also need to be aligned and calibrated. Referring to the embodiment shown in fig. 3, as shown in fig. 4, 9 photosensitive elements in the first photosensitive element array 41 form a main photosensitive area to collect echo signals of a sub-area, and one auxiliary photosensitive area 421 in the second photosensitive element array 42 is correspondingly used to collect echo signals of the same sub-area, where the auxiliary photosensitive area 421 may include at least one second photosensitive element 422, and the size of the second photosensitive element is larger than that of the first photosensitive element. In one embodiment, assuming that the first photosensitive element 413 in the main photosensitive area collects echo signals reflected from a conventional object, the first photosensitive element 412 collects echo signals reflected from snow. The second photosensitive element 422 in the auxiliary photosensitive area can collect the echo signal reflected by the same snowflake.
However, in practical applications, on the one hand, the plurality of first photosensitive elements in the main photosensitive area 411 may not necessarily receive the reflected light spots; on the other hand, the emitted light spot may irradiate the object behind the spatial suspension through the spatial suspension when the emitted light spot irradiates the spatial suspension, and particularly the auxiliary radar module emits a large light spot beam, when the spatial suspension is detected, the spatial suspension occupies only a small part of the large light spot beam, and other most light beams may irradiate other objects such as a conventional object positioned behind the spatial suspension or other spatial suspensions behind the spatial suspension through the spatial suspension, so that the distance data which each photosensitive element can output is not only one, but may be plural when the multi-echo condition occurs. For example, as shown in fig. 4, the second photosensitive element 422 may also receive an echo signal of a conventional object to output a distance data. Therefore, in the scheme of the application, the single echo or multiple echoes are not required to be considered, only a group of data to be analyzed is formed by the first distance output by the corresponding main photosensitive area and the second distance output by the auxiliary photosensitive area, analysis and judgment are carried out one by one, and finally, the first distance data with the target of space suspension is eliminated.
The data processing circuit can accurately acquire the main radar data and the auxiliary radar data corresponding to each sub-area to form a group of data to be processed by carrying out alignment calibration setting on the position relation of the projection of the light spots to the space and the photosensitive element array in advance, and further carries out subsequent processing on the data.
In some embodiments, to optimize the system structure, the main collector and the auxiliary collector can be configured to be integrated together, and share one photosensitive element array, so as to realize adjustment of the photosensitive area by regulating and controlling the number of activated photosensitive elements. The primary emitter includes a first light source array comprised of a plurality of first light sources and the secondary emitter includes a second light source array comprised of a plurality of second light sources. Preferably, the number of second light sources is smaller than the number of first light sources, i.e. each large spot beam may cover at least two small spot beams. In one embodiment, the setting of the spot size, e.g. the light emission aperture, may be achieved by configuring parameters of the array of light sources, i.e. the light emission aperture of the first light source is larger than the light emission aperture of the second light source. In another embodiment, the second light source array may be disposed out of focus with the second emission lens such that the light spot emitted by the second light source diverges through the second emission lens to expand the size of the light spot beam.
In other embodiments, the main emitter and the auxiliary emitter can be selectively configured to be integrated into the same emitter, and two collectors are correspondingly provided as the main collector and the auxiliary collector, that is, the first light source and the second light source are array light source chips integrated into the same semiconductor substrate, and the size of the emission light spot is changed by configuring different light emitting apertures of the light sources. When the main radar works, the first light source and the main collector are regulated and controlled to work to generate detection data, and when the auxiliary radar works, the second light source and the auxiliary collector are regulated and controlled to work to generate detection data.
In still other embodiments, the main transmitter and the auxiliary transmitter can be optionally configured to be integrated into the same transmitter, the main collector and the auxiliary collector are integrated into one collector, and the main radar and the auxiliary radar take turns to measure the detection space.
In some embodiments, a scanning unit may also be included in the primary and secondary radar modules. The same or different scanning units can be respectively configured in the main radar module and the auxiliary radar module, or the main radar module and the auxiliary radar module can share the scanning units.
(IV) anti-dry anti-laser radar measurement method
Fig. 5 is a schematic process diagram of a measurement method according to an embodiment of the application, and the method is applied to a data processing circuit. The method specifically comprises the following steps:
s501: receiving first distances corresponding to a plurality of small light spot beams in a first detection range output by a main radar module;
S502: receiving second distances and second echo intensities corresponding to a plurality of large-spot light beams in a second detection range output by an auxiliary radar module;
s503: calculating a second target reflectivity corresponding to each large-spot light beam according to the second distance and the second echo intensity;
s504: judging whether the target corresponding to each small light spot beam is a space suspension or not by utilizing the first distance, the second distance and the second target reflectivity;
S505: and optimizing the first distance according to the judgment result to obtain the distance information of the target corresponding to each small light spot beam in the first detection range.
Wherein the first detection range is larger than the second detection range; the main radar module and the auxiliary radar module respectively emit a plurality of small light spot beams and one large light spot beam to each sub-area in the detection space.
In some embodiments, the second target reflectivity is compared to a second reflectivity cut-off threshold and the distance deviation of the first distance and the second distance is compared to a distance threshold, the target being a regular object when the second target reflectivity is greater than the second reflectivity cut-off threshold and the distance deviation is not greater than the distance threshold. In other cases, the target is a spatial suspension, and further, the first distance of the target is eliminated. The second reflectivity cut-off threshold is a reflectivity cut-off threshold of the auxiliary radar module set through pre-calibration.
In some embodiments, further comprising: receiving first echo intensities corresponding to a plurality of small light spot beams in a first detection range output by a main radar module; if the first distance is greater than the upper measurement limit of the second detection range, the method further comprises: calculating a first target reflectivity corresponding to the small light spot beam according to the first distance and the first echo intensity; judging whether a target corresponding to the small light spot beam is a space suspension or not by utilizing the first target reflectivity and the first reflectivity cut-off threshold; and optimizing the first distance according to the judgment result to obtain the distance information of the target corresponding to the small light spot beam in the first detection range. Typically, the target is a conventional object when the first target reflectivity is greater than the first reflectivity cut-off threshold and is a spatial suspension when not greater than the first reflectivity cut-off threshold. Further, if the target is a spatial suspension, the first distance data of the target is eliminated.
In some embodiments, when the first distance is not greater than the upper measurement limit of the second detection range, if the first distance corresponding to the small spot beam is valid and the second distance corresponding to the large spot beam and the second echo intensity are invalid, it is determined that the target corresponding to the small spot beam is a spatial suspension. Further, the first distance of the target is eliminated.
As shown in fig. 6, the main photosensitive area outputs 4 first distances and the auxiliary photosensitive area outputs 2 second distances are illustrated as an example. The main histogram 61 includes 4 echo pulses 610, 611, 612 and 613, which respectively correspond to the first distance d10、d11、d12、d13; the secondary histogram includes 2 echo pulses 620, 621 corresponding to a second distance d20、d21 (shown in dashed lines as a hover echo pulse and solid lines as a normal object echo pulse), respectively. The 4 first distances of the primary radar output need to be analyzed separately to determine if the target is spatially suspended.
The first distance data corresponding to the echo pulses 610, 611 and 612 are all located in the second detection range of the auxiliary radar module, and the comparison analysis is required to be performed on the data output by the auxiliary radar module, so as to identify whether the target corresponding to each first distance output by the main radar module is a spatial suspension. For example, the echo pulse 610 is analyzed with the echo pulses 620 and 621, and first, the echo pulse 620 is screened out according to the condition that the distance deviation is not greater than the distance threshold, and it is further calculated that the target reflectivity corresponding to the echo pulse 620 is smaller than the second reflectivity cut-off threshold, and then the second distance d10 needs to be eliminated. Similarly, echo pulse 611 is analyzed with echo pulses 620 and 621, respectively, and if the distance deviation does not satisfy the condition, then the echo pulse 611 needs to be removed; the distance deviation between the echo pulse 612 and the echo pulse 621 is not greater than the distance threshold, and the second target reflectivity corresponding to the echo pulse 621 is greater than the second reflectivity cut-off threshold, then the second distance d21 is reserved. And obtaining the distance information of the targets in the first detection range after analyzing one by one.
In the scheme of the application, the distance deviation and the reflectivity are unavoidably compared, but the execution sequence of the two conditions is not strictly limited, and in the embodiment, only schematic description is made, the second target reflectivity can be calculated first to determine whether the object is a space suspension or not, and then the distance deviation is used for judging whether each first distance in the main radar module needs to be reserved or removed.
The first distance d13 corresponding to the echo pulse 613 is greater than the upper measurement limit of the second detection range, which may theoretically be considered as the distance of a conventional object, but in view of some special scenarios, the data may also be analyzed to determine whether the target is a spatial suspension. In some embodiments, a first target reflectivity may be calculated based on a corresponding first echo intensity for the first distance d13, and whether a conventional object or a spatial suspension is analyzed based on the first target reflectivity. Typically, a first reflectivity cut-off threshold of the main radar module is set, and if the target reflects more than the first reflectivity cut-off threshold, the target is reserved, otherwise, the target is rejected. It will be appreciated that the first and second reflectivity cut-off thresholds may be the same or different, and may be specifically set according to parameters of the secondary and primary radar modules.
In some embodiments, the first distance data corresponding to the echo pulses 610, 611, 612 are all located in the second detection range of the auxiliary radar module, that is, the first distance data is not greater than the upper measurement limit of the second detection range, if the data processing circuit determines that the second distance and the second echo intensity output by the auxiliary radar module are both invalid values (the signal-to-noise ratio is too low to cause invalidation) when the read data are, and the first distance of the data of the main radar module is valid, it may be determined that the targets corresponding to the first distances are all spatial suspended matters directly, and the first distances corresponding to the targets are removed.
In some embodiments, after the first distance is optimized according to the determination result to obtain the distance information of the target corresponding to each small light spot beam in the first detection range, in the actual detection process, special situations such as particles of the spatial suspension or some random noise exist, which results in some noise points in the optimized distance information of the target. Further, the distance information of the target in the first detection range may be filtered to obtain the distance information of the target without noise. Typically, the distance information may be filtered based on spatial continuity filtering, taking into account the continuity of the target and the isolation of the noise.
Furthermore, the application also provides a mobile platform, which comprises the laser radar system
And a platform body, wherein the lidar system is mounted on the platform body.
It will be appreciated that the method provided by the present application is applied to the lidar described above, and the specific implementation of the method steps has been described in detail in the foregoing embodiments, and will not be repeated here.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.