Disclosure of Invention
The invention provides a method for acquiring the motion parameters of a lead train in real time so as to track the lead train and cooperatively control the whole train grouping when the lead train establishes the grouping operation, thereby guaranteeing the operation safety of the virtual linkage train.
The invention provides a perception system of a motion state of a virtual train, which comprises the following components: the integrated host and the functional module are electrically connected with the current train, and the RFID tag is arranged on the lead train; the functional module includes: the system comprises a camera module, a radar module, an inertial measurement module and an RFID antenna;
the camera module is used for identifying the lead train;
The radar module is used for measuring the real-time speed of the current train; measuring the relative distance when the train is connected with the front sequence train to establish the marshalling operation; measuring the relative speed of the lead train and the current train when the lead train is linked with the lead train to establish a marshalling operation;
the inertia measurement module is used for acquiring the radial acceleration of the current train;
The integrated host is used for establishing marshalling operation with the front sequence train linkage based on the real-time speed and the radial acceleration of the current train; and based on the relative distance between the front train and the front train when the front train is linked and the relative speed between the front train and the current train, the front train is tracked by combining the RFID antenna with the RFID tag installed on the front train.
Optionally, the camera module includes: a long-focus camera and a short-focus camera;
the long-focus camera is used for acquiring a front sequence train target exceeding the preset distance of the train;
The short-focus camera is used for compensating the visual field change caused by the railway curve when the short-focus camera is connected with the front sequence vehicle to establish the marshalling operation.
Optionally, the radar module includes: secondary radar, lidar, millimeter wave radar:
The laser radar is used for determining the speed state of the current train and measuring the real-time speed of the current train when the speed state is a low-speed state; continuously acquiring the real-time distance between the lead train and the current train according to the frequency of 10Hz when the lead train is linked with the lead train to establish marshalling operation;
The millimeter wave radar is used for measuring the relative speed of the lead train and the own vehicle and the absolute speed of the own vehicle;
the secondary radar is used for determining target capturing and ranging of the same-track lead train beyond a preset distance based on the RFID tag and the RFID antenna.
Optionally, the integrated host includes: a CPU processor, a GPU processor and a communication interface board;
The CPU processor is used for predicting the motion state of the lead train based on the relative distance and the relative speed by adopting a preset algorithm to obtain a prediction result, and tracking the lead train according to the prediction result;
the GPU processor is used for matching the image information acquired by the camera module and the motion state data acquired by the radar module;
The communication interface board is used for providing a data interface for the functional unit.
Optionally, the laser radar is specifically configured to: judging whether the real-time speed of the current train is less than 10km/h, if so, determining that the speed state of the current train is a low-speed state.
Optionally, the laser radar is further specifically configured to:
and determining the real-time speed of the current train by acquiring the relative position change data of the current train in the low-speed state in a preset measurement period.
Optionally, the laser radar is further specifically configured to: when the train is connected with the front sequence train to establish marshalling operation, the ranging information of the front sequence train and the current train is continuously acquired according to the preset acquisition frequency.
Optionally, the preset acquisition frequency is 10Hz.
Optionally, the CPU processor is specifically configured to: and tracking the lead train by adopting a single-target tracking mode according to the prediction result.
Optionally, the preset algorithm is a kalman filter algorithm.
From the above technical scheme, the invention has the following advantages:
The invention provides a perception system of a motion state of a virtual train, which comprises the following components: the integrated host and the functional module are electrically connected with the current train, and the RFID tag is arranged on the lead train; the functional module includes: the system comprises a camera module, a radar module, an inertial measurement module and an RFID antenna; the camera module is used for identifying the lead train; the radar module is used for measuring the real-time speed of the current train; measuring the relative distance when the train is connected with the front sequence train to establish the marshalling operation; measuring the relative speed of the lead train and the current train when the lead train is linked with the lead train to establish a marshalling operation; the inertia measurement module is used for acquiring the radial acceleration of the current train; the integrated host is used for establishing marshalling operation with the front sequence train linkage based on the real-time speed and the radial acceleration of the current train; and based on the relative distance between the front train and the front train when the front train is linked and the relative speed between the front train and the current train, the front train is tracked by combining the RFID antenna with the RFID tag installed on the front train.
Based on the radar module and the inertia measurement module, the motion parameters of the lead trains in the virtual train are acquired in real time, then the lead trains are tracked under the assistance of the integrated host, and the whole train grouping is cooperatively controlled when the lead trains are in grouping operation, so that the safety of the virtual train operation is ensured, and the efficiency of the virtual train operation is improved.
Detailed Description
The embodiment of the invention provides a perception system of a motion state of a virtual train, which is used for acquiring motion parameters of a preceding train in real time so as to track the preceding train, cooperatively controlling the whole train grouping when the preceding train is in grouping operation, and guaranteeing the operation safety of the virtual train.
In order to make the objects, features and advantages of the present invention more comprehensible, the technical solutions in the embodiments of the present invention are described in detail below with reference to the accompanying drawings, and it is apparent that the embodiments described below are only some embodiments of the present invention, but not all embodiments of the present invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to fig. 1, fig. 1 is an overall architecture diagram of a perception system for motion state of a virtual train, which is provided in an embodiment of the present application, and the system includes: the integrated host 1 and the functional module 2 are electrically connected with the current train, and the RFID tag is arranged on the lead train; the functional module 2 includes: a camera module 21, a radar module 24, an inertial measurement module 22 and an RFID antenna 23;
The camera module 21 is used for identifying the lead train;
The radar module 24 is used for measuring the real-time speed of the current train; measuring the relative distance when the train is connected with the front sequence train to establish the marshalling operation; measuring the relative speed of the lead train and the current train when the lead train is linked with the lead train to establish a marshalling operation;
The inertial measurement module 22 is configured to obtain a radial acceleration of the current train;
The integrated host 1 is configured to establish a marshalling operation with the pre-train linkage based on the real-time speed and the radial acceleration of the current train; and based on the relative distance between the front train and the front train when the front train is linked to establish the marshalling operation and the relative speed between the front train and the current train, the RFID antenna 23 is combined with the RFID tag installed on the front train to track the front train.
It should be noted that the virtual coupling means two or more train groups, and the near-distance tracking cooperative operation is realized by a wireless train communication mode, and a virtual (without physical coupler connection) large train group operation is formed. The virtual linkage is a new technical form for improving the flexibility of rail transportation and coping with tidal passenger flow and large-station express operation.
The inertial measurement unit 22 (IMU) is a very important device that can measure and calculate acceleration and angular velocity of an object in three different directions. Accordingly, IMUs are widely used in a variety of fields including navigation, motion control, aerospace, and robotics, among others. Specifically, the IMU may calculate position, velocity, and attitude information of an object by measuring acceleration and angular velocity of the object. It is typically composed of various sensors such as accelerometers, gyroscopes, and magnetometers that can detect the state of motion of an object through minute vibrations and rotations. By calculating and analyzing the motion states, the IMU can calculate the position, speed and gesture information of the object, thereby realizing applications such as navigation and control.
The perception system of the motion state of the virtual train provided by the embodiment of the invention can continuously and reliably autonomously perceive the motion parameters of the lead train and the current train through the radar module 24 and the inertia measurement module 22 in the running process of the virtual train without depending on the signal system and the network control system of the current train, and the perceived result of the system can be used for cooperative control and collision protection of train formation and support virtual train formation running.
Further, the camera module 21 includes: a tele camera 211 and a short-focus camera 212;
The long-focus camera 211 is used for acquiring a front sequence train target exceeding a preset distance of a train;
The short-focus camera 212 is used for compensating the visual field change caused by the railway curve when the train is connected with the front sequence train to establish the marshalling operation.
In the embodiment of the invention, the camera module 21 is formed by adopting a combination mode of the long-focus camera 211 and the short-focus camera 212, and the acquisition of the lead train beyond the preset distance is acquired through the long-focus camera 211, so that the target acquisition of the lead train at a longer distance is realized; the short-focus camera 212 is utilized to compensate the visual field change caused by the railway curve when the train is connected with the front sequence car to establish the marshalling operation, namely blind compensation, so that the curve adaptability is improved. And the characteristics of different sensors in the system are combined, and the target matching and stable tracking locking of the lead train are realized by adopting a mode of distance grading and multistage gradual fusion.
Further, the radar module 24 includes: secondary radar 241, laser radar 242, millimeter wave radar 243:
The lidar 242 is configured to determine a speed state of the current train, and measure a real-time speed of the current train when the speed state is a low speed state; continuously acquiring the real-time distance between the lead train and the current train according to the frequency of 10Hz when the lead train is linked with the lead train to establish marshalling operation;
millimeter wave radar 243 for measuring the relative speed of the preceding train and the own vehicle and the absolute speed of the own vehicle;
The secondary radar 241 is configured to determine, based on the RFID tag and the RFID antenna 23, target capturing and ranging of the co-track lead train outside a preset distance.
Laser radar 242 is a device for measuring distance and detecting an object based on laser technology. It calculates the distance between the measured object and the radar by emitting a pulsed laser and measuring the time required for the laser beam to bounce back. The lidar 242 has advantages of accuracy, high density data, high speed imaging capability, and adaptability to various environments.
The millimeter wave radar 243 is a radar technology based on millimeter wave high frequency electromagnetic waves. It can utilize radio frequency outside microwave frequency band, and can implement radar signal transmission and detection under the condition of high bandwidth and high resolution. Compared with the conventional radar technology, the millimeter wave radar 243 has higher frequency and shorter wavelength, thus having higher resolution and better penetrating power, and can more accurately detect the information of the position, speed, direction and the like of the moving object.
The secondary radar 241 is radar technology based on radio or electromagnetic wave signals such as millimeter waves or infrared rays. Unlike conventional radars, the secondary radar 241 does not directly transmit radar signals, but uses other transmission media as a detection signal source to measure and analyze interactions with other objects or obstacles, thereby implementing ranging, image reconstruction, environment sensing, and other functions.
The perception system of the virtual linkage train motion state in the embodiment of the invention comprises the following steps when the motion state measurement of the pre-sequence train is realized:
(1) And (5) target identification. The camera module 21 formed by the long-focus camera 211 and the short-focus camera 212 is used for detecting and identifying targets of the front sequence vehicles to be virtually linked, and detecting whether the targets are in the current track or not, wherein the targets comprise detection, distance (long-short-focus binocular vision depth estimated value) and whether the targets are in the current track or not; and identifying the front sequence vehicle type and the vehicle number in front of the vehicle head.
(2) And (5) target matching. In the process that the virtual train is gradually approaching, the sensors of various types sequentially detect the targets of the train in the front sequence. After the long-short-focus binocular camera module 21 stably detects and identifies the lead train, performing target matching of result level fusion with the secondary radar 241, wherein the matching basis is target distance and same track operation; after the laser radar 242 detects the lead train, the lead train is matched with a target of the long-short-focus binocular camera module 21 in the pixel level fusion in space, so that accurate target distance and position information are obtained; after the millimeter wave radar 243 detects the lead train, the result level target matching is carried out on the lead train and the fusion target obtained by the camera and the laser radar 242 in space, so as to obtain the relative speed of the target; finally, each sensor detects the target train and realizes fusion.
(3) Target tracking lock. The sensing system continuously detects the target train, the detection result updating period is 100ms, a single target tracking mode (STT) is adopted to record the track of the lead train, a Kalman Filtering (KF) algorithm is adopted to predict the motion state of the lead train, a laser radar 242 provides accurate distance input, a millimeter wave radar 243 provides relative speed input, and an IMU provides own vehicle acceleration correction input.
The laser radar 242 is used for carrying out high-precision distance measurement and accurate speed measurement in a low-speed state when the train is hung in a linked mode to establish a marshalling operation;
(4) The relative speed measurement of the front sequence vehicle and the own vehicle and the measurement of the own absolute speed of the own vehicle are realized through the millimeter wave radar 243;
(5) Target capturing and ranging of the long-distance co-track lead train are achieved through the secondary radar 241 and the RFID. 12
Further, the integrated host 1 includes: a CPU processor 11, a GPU processor 12, and a communication interface board 13;
The CPU processor 11 is configured to predict a motion state of the lead train based on the relative distance and the relative speed by using a preset algorithm, obtain a prediction result, and track the lead train according to the prediction result;
the GPU processor 12 is configured to match the image information acquired by the camera module 21 with the motion state data acquired by the radar module 24;
The communication interface board 13 is configured to provide a data interface for the functional unit.
In the embodiment of the invention, the integrated embedded platform host integrates main functional units such as a CPU processor 11, a GPU processor 12, a communication interface board 13 and the like, and the CPU processor 11 is used for processing laser radar 242 point cloud clustering and ranging, speed measurement of millimeter wave radar 243, secondary radar 241 communication ranging, IMU pose and acceleration measurement and RFID tag information; the GPU processor 12 performs machine vision image recognition and sensor information fusion; the communication interface board 13 provides a data interface for the respective functional units and sensors and is finally converted into data information transmitted through the ethernet.
Further, the laser radar 242 is specifically configured to: judging whether the real-time speed of the current train is less than 10km/h, if so, determining that the speed state of the current train is a low-speed state.
Further, the laser radar 242 is specifically configured to:
and determining the real-time speed of the current train by acquiring the relative position change data of the current train in the low-speed state in a preset measurement period.
In an alternative embodiment, the lidar 242 is also specifically configured to: when the train is connected with the front sequence train to establish marshalling operation, the ranging information of the front sequence train and the current train is continuously acquired according to the preset acquisition frequency.
Specifically, aiming at the characteristics of different radars, the perception system of the motion state of the virtual train realizes relative distance measurement, absolute position positioning, absolute speed measurement and relative speed measurement by the following methods:
(1) Relative distance measurement: after finishing identification, same orbit judgment and locking of the front sequence vehicle target, starting to continuously measure the relative distance, respectively obtaining distance values with different precision through a secondary radar 241, a camera module 21, a laser radar 242 and a millimeter wave radar 243, then adopting a secondary radar 241 ranging result with highest reliability as a reference standard, correcting by taking a high-precision ranging value of the laser radar 242 as a correction basis, and judging the validity of the correction result by utilizing the measured distance values of the camera module 21 and the millimeter wave radar 243 after correction. Thereby obtaining a relative distance measurement result with the measurement precision that the relative distance between the front sequence vehicle and the vehicle is not more than 0.1 m.
(2) Absolute position location: performing sparse point cloud mapping on the whole line by using a laser radar 242, identifying a hundred meter mark beside a track by using a camera module 21, and performing position correction by combining with an RFID tag of the current train to quickly determine a map retrieval range; and then, the laser radar 242 point cloud data acquired in real time are subjected to inertial measurement module 22 to correct the movement data and vibration data of the current train, and are quickly matched with a pre-acquired sparse map after correction, so that the accurate position of the current train in the whole line is determined. Thereby obtaining absolute position location with accuracy not greater than 0.5 m.
(3) Absolute velocity measurement: in a high-speed scene (generally 10km/h or more), the characteristics that most targets in the emission angle range of the millimeter wave radar 243 are stationary targets are utilized, the weight number in the speed measurement result of each target is taken, and the relative speed of the current train on a ground stationary object is determined, so that the absolute speed of the current train is indirectly measured; in a low-speed scene, the laser radar 242 is used for measuring the relative position change of the sleeper at a stable period of 100ms, so that high-precision speed information is obtained. By adopting the mode of combining the millimeter wave radar 243 and the laser radar 242, the high-precision measurement of the absolute speed of the current train in the high-speed and low-speed intervals is realized, and the measurement precision can reach 0.1m/s.
(4) Relative velocity measurement: after the identification of the targets of the front sequence vehicles and the judgment and locking of the same track are completed, the measurement of the relative speed of the front sequence vehicles can be continuously carried out, the relative speed of the front sequence vehicles is directly measured by adopting a millimeter wave radar 243 in the measurement process, the reference speed is obtained by adopting ranging information of a stable period of a secondary radar 241 and a laser radar 242, and when the difference of the relative speeds measured by the millimeter wave radar is less than 1%, the speed measurement of the millimeter wave radar 243 is considered to be effective; under the condition of small relative speed with the lead train, namely stable running, the speed measurement precision of the millimeter wave radar 243 is seriously reduced, and the laser radar 242 is adopted to calculate and calculate the relative speed between the current train and the lead train according to the high-precision ranging information continuously output by the preset acquisition frequency. Therefore, under the condition that the relative speed of the current train and the front train is smaller, the speed measurement precision of 0.1m/s can still be achieved.
Further, the preset acquisition frequency is 10Hz.
Further, the CPU processor 11 is specifically configured to: and tracking the lead train by adopting a single-target tracking mode according to the prediction result.
It should be noted that single-object tracking refers to tracking a single object in a video or image sequence. The aim is to determine the position of the object in successive frames and, where possible, to estimate its speed and direction. Common single-target tracking methods include: color-based tracking, shape-based tracking, motion-based tracking, feature-based tracking, and deep learning-based tracking.
Further, the preset algorithm is a Kalman filtering algorithm.
It should be noted that the kalman filter algorithm is an algorithm for estimating a system state, and is based on a linear dynamic system model and gaussian noise assumption. It can iteratively update the estimates of the system state by measuring the data and the system model and estimate the uncertainty of the system state. The method is widely applied to the fields of control systems, navigation, robots, signal processing and the like.
In the embodiment of the invention, the operation state of the lead train can be better judged by combining the Kalman filtering algorithm on the basis of the single-target tracking method, thereby improving the precision and stability of the current train when tracking the lead train.
The system designed by the invention mainly works in the following four operation scenes, and the corresponding strategies in each scene are as follows:
(1) In an independent running state, the current train is a first train of a virtual linkage grouping or a front train without a front train sequence, and at the moment, the current train does not establish a virtual linkage running requirement with the front train sequence; the system designed by the embodiment of the invention is in an obstacle detection mode, namely, only the obstacle with collision risk in the limit in front of the train operation is detected, and the system is used as an active anti-collision system, and each sensor carries out indiscriminate detection on the front target;
(2) When a train receives a command for establishing a linkage, the system designed by the embodiment of the invention is switched to a linkage mode, and starts to utilize a secondary radar 241 to confirm whether the distance between a front train and a running track meet the requirement for establishing virtual linkage (the distance between two workshops is smaller than 500m and the running track is the same), if the requirements can be met, the current train is allowed to further approach the front train, and the targets of the front train are detected and matched sequentially through a long-focus camera 211, a short-focus camera 212, a laser radar 242 and a millimeter wave radar 243, and after all sensors capture targets of the front train, the system enters a target locking state to inform the train of completing linkage establishment;
(3) In the running process of establishing a virtual coupling and marshalling of a current train, the perception system of the motion state of the virtual coupling and marshalling train continuously carries out kinematic parameter measurement on a locked front sequence train target, wherein the measurement contents comprise, but are not limited to, the relative distance and relative speed between a front train and the train, the absolute position and absolute speed of the train on a line and the like, and the real time coupling and marshalling train is multicast to a train network control system and a signal system through an Ethernet TRDP protocol and is used for cooperative control in the running process of the virtual coupling and marshalling train; when the target is lost, the virtual train is informed of ending the train operation, and the braking guiding is safe;
(4) When the train is normally linked and unbinding, the current train sends a signal to inform the perception system of the motion state of the virtual linked train, namely the virtual linked train can enter the linked and unbinding state, and as the distance between the current train and the lead train increases, the millimeter wave radar 243, the laser radar 242, the short-focus camera 212 and the long-focus camera 211 sequentially exceed the target detection range, and in the process of gradually keeping away from the lead train, the secondary radar 241 is adopted to monitor the train distance, so that each sensor is stably switched in the boundary interval of the preset detection range, and the target loss caused during unbinding is avoided, so that the emergency braking of guiding safety is caused.
The sensing system for the motion state of the virtual articulated train provided by the embodiment of the application comprises the following components: the integrated host 1 and the functional module 2 are electrically connected with the current train, and the RFID tag is arranged on the lead train; the functional module 2 includes: a camera module 21, a radar module 24, an inertial measurement module 22 and an RFID antenna 23; the camera module 21 is used for identifying the lead train; the radar module 24 is used for measuring the real-time speed of the current train; measuring the relative distance when the train is connected with the front sequence train to establish the marshalling operation; measuring the relative speed of the lead train and the current train when the lead train is linked with the lead train to establish a marshalling operation; the inertial measurement module 22 is configured to obtain a radial acceleration of the current train; the integrated host 1 is configured to establish a marshalling operation with the pre-train linkage based on the real-time speed and the radial acceleration of the current train; and based on the relative distance between the front train and the front train when the front train is linked to establish the marshalling operation and the relative speed between the front train and the current train, the RFID antenna 23 is combined with the RFID tag installed on the front train to track the front train.
Based on the radar module 24 and the inertia measurement module 22, the motion parameters of the lead trains in the virtual train are acquired in real time, then the lead trains are tracked with the aid of the integrated host 1, and the whole train grouping is cooperatively controlled when the lead trains are in grouping operation, so that the safety of the virtual train operation is ensured, and the efficiency of the virtual train operation is improved.
In the several embodiments provided by the present application, it should be understood that the disclosed apparatus and method may be implemented in other manners. For example, the apparatus embodiments described above are merely illustrative, e.g., the division of the units is merely a logical function division, and there may be additional divisions when actually implemented, e.g., multiple units or components may be combined or integrated into another system, or some features may be omitted or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be an indirect coupling or communication connection via some interfaces, devices or units, which may be in electrical, mechanical or other form.
The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional unit in the embodiments of the present application may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units may be implemented in hardware or in software functional units.
The integrated units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application may be embodied in essence or a part contributing to the prior art or all or part of the technical solution in the form of a software product stored in a storage medium, comprising several instructions for executing all or part of the steps of the method according to the embodiments of the present application by means of a computer device (which may be a personal computer, a server, or a network device, etc.). And the aforementioned storage medium includes: u disk, mobile hard disk, read-Only Memory (ROM), random access Memory (Random Access Memory, RAM), magnetic disk or optical disk, etc.
The above embodiments are only for illustrating the technical solution of the present application, and not for limiting the same; although the application 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 application.