Method and device for determining a set acceleration trajectory of an adaptive vehicle speed regulator of an autonomous vehicleTechnical Field
The present invention claims priority from french application 2011872 filed 11/19 2020, the contents of which (text, figures and claims) are incorporated herein by reference. The present invention relates to the field of driving assistance systems for autonomous vehicles. In particular, the present invention relates to a determination method and a determination apparatus for determining a set acceleration trajectory of an adaptive vehicle speed regulator of an autonomous vehicle called a host vehicle to dock (accoloter) a vehicle traveling in front of the host vehicle called a preceding vehicle.
Background
"vehicle" is understood to mean any type of vehicle, for example a motor vehicle, a motor bicycle, a motorcycle, a storage robot in a warehouse, etc. "autonomous driving" of an "autonomous vehicle" is understood to mean any method capable of assisting the driving of the vehicle. The method may thus consist in partly or fully guiding the vehicle or providing any type of assistance to the natural person driving the vehicle. The method thus covers autonomous driving from rank 0 to rank 5 in an OICA (i.e. "Organisation International des Constructeurs Automobiles (international society of automotive manufacturers)") rank table.
The method capable of assisting the driving of the vehicle is also called an ADAS (acronym of english "Advanced driver-assistance systems"), an ADAS system, or a driving assistance system. The vehicle speed regulator RVV is a known ADAS system for regulating the speed of a vehicle according to a given speed set point (referred to as a set speed). When the set speed is changed, a plurality of vehicle speed trajectories (time evolution of the vehicle speed) for achieving the set speed are thereby possible. Following the vehicle speed trajectory, vehicle acceleration, and thus vehicle dynamic behavior, is determined. Some regulators are also determined by taking into account the acceleration, in order to typically avoid excessively strong accelerations (positive or negative) that cause discomfort to the vehicle occupants.
An adaptive vehicle speed regulator or ACC (acronym for english "Adaptive Cruise Control (adaptive cruise control)") is also a known ADAS system for regulating the speed of a vehicle and the inter-vehicle time, which characterizes the duration of a forward or rear pass of two vehicles that are connected separately on the same pass lane. For example, the inter-vehicle time is a predetermined parameter predetermined by the driver or predetermined parameter predetermined by default according to the current regulatory recommendation (e.g., 2 seconds). The predetermined inter-vehicle time is also referred to as a target time hereinafter.
The apparatus of the vehicle including the ACC function can acquire longitudinal dynamic characteristics of the own vehicle and the traveling preceding vehicle. The means include, for example, cameras, radars, lidars, etc. The device is capable of detecting the arrival of a traveling preceding vehicle. This is the case, for example, when a certain vehicle turns onto the same traffic lane as the own vehicle, or when the own vehicle catches up with a vehicle traveling ahead of the own vehicle and traveling slower than the own vehicle on the same traffic lane. In particular, the speed and acceleration of the own vehicle, the speed and acceleration of the traveling preceding vehicle, and the distance between the own vehicle and the traveling preceding vehicle (referred to as inter-vehicle distance) are measured. By using vehicle speed information, the inter-vehicle time is derived from the inter-vehicle distance, and vice versa.
When a preceding vehicle is detected, the speed of the own vehicle needs to be adapted in order to follow the target time (or target distance). Typically, the speed of the traveling preceding vehicle is measured and employed as the target speed. The speed regulator adapts the speed and thus the acceleration of the own vehicle to dock with the traveling preceding vehicle. By "docking" is understood adapting the speed of the own vehicle over a determined period of time to reach the speed of the traveling preceding vehicle while following the inter-vehicle time at the end of the determined period of time. The vehicle speed regulator determines only the speed trajectory or acceleration trajectory to be followed by the vehicle. This determines the always identical docking behavior (longitudinal dynamic behavior of the self-vehicle). The driver may wish to classify the interfacing behavior (e.g., suspension behavior or automatic gearbox behavior) as they wish to classify, depending on his mood, depending on the vehicle occupancy (driver alone, driver with his child, etc.), or depending on his needs.
Disclosure of Invention
The object of the present invention is to solve the above-mentioned problems, in particular to calculate acceleration trajectories for docking a preceding vehicle according to a desired behavior by means of a vehicle speed regulator.
To this end, a first aspect of the invention relates to a determination method for determining a set acceleration trajectory of an adaptive vehicle speed regulator of an autonomous vehicle, referred to as a self-vehicle, to dock with a vehicle travelling in front of the self-vehicle, referred to as a preceding vehicle. The method comprises the following steps:
-determining a classification parameter k, said classification parameter characterizing a docking behaviour;
-detecting said traveling preceding vehicle on the same traffic lane as said own vehicle, said detecting determining an initial moment;
-determining a target speed Vc, which characterizes the speed that the own vehicle needs to have at a predetermined end time tf when the own vehicle has docked the travelling preceding vehicle;
-determining a distance deviation D between the measured inter-vehicle distance D0 and a target inter-vehicle distance Dc, the target inter-vehicle distance being determined based on the target speed and a predetermined target inter-vehicle time;
-determining a set acceleration trajectory based on the distance deviation D, the end time tf and the classification parameter k.
Different docking behaviors are possible due to the present invention. For example, one of the driving modes is a rapid docking, which is characterized by a strong longitudinal vehicle dynamics (i.e. a strong acceleration followed by a strong deceleration in order to adapt the inter-vehicle distance as quickly as possible). Another driving mode is a slow and dynamically comfortable docking that is characterized by weak longitudinal vehicle dynamics. The intermediate driving mode is a psychologically comfortable dock characterized by a more intense longitudinal vehicle dynamics at the beginning of the maneuver at an initial time than at the end of the maneuver at an end time. In the last case, the driver perceives good detection of the target vehicle at the beginning of the maneuver.
The numerical value is assigned to the classification parameter k as a function of the selection of the driving mode by the driver. Thus, by setting the duration for docking (i.e. the end time tf) to know the distance deviation to be followed, and by using the classification parameter k, the set acceleration trajectory is determined. The trajectory is parameterized by the classification parameter k. Thus, by a single parameter, the self-vehicle is classified differently from the docking of the traveling preceding vehicle. For example, by multiplying the steering time tf by the classification parameter k, the trajectory lasts more or less time, whereby the vehicle has different longitudinal dynamics depending on the value of the classification parameter k, since the speed regulator regulates the speed of the own vehicle according to the set acceleration trajectory. Having a single classification parameter k may also simplify the adjustment of the regulator. In fact, when the regulator is adjusted for normal behaviour, it is only necessary to choose two further digital values for the classification parameter k in order to have two further different interfacing behaviour.
Advantageously, the determination of the set acceleration trajectory a (t) is obtained based on the following equation:
where t is time and, at the initial time, t=0.
Thus, only a few parameters (D, tf and k) are required to calculate the set acceleration trajectory according to the classification of the desired driving pattern. The classification parameter k adjusts the end time to give an equivalent end time. When the classification parameter k is smaller than 1, the equivalent end time is smaller. Conversely, when the classification parameter k is greater than 1, the equivalent time is greater.
Advantageously, the method further comprises determining that the data is present in the dataDetermining step of a speed deviation V between a measured speed V0 of the own vehicle and a target speed Vc for the vehicle based on the equationA determining step of determining the end time tf.
The end time tf and thus the docking maneuver duration are thus simply calculated and determined based on the inter-vehicle distance measured at the initial time, the speed of the own vehicle at the initial time, the target inter-vehicle distance and the target speed. The target inter-vehicle distance is determined based on the target inter-vehicle time and the target speed. In one mode of operation, the target speed is a speed of the traveling preceding vehicle at the initial time. In another mode of operation, the target speed is a speed determined based on a speed of the traveling preceding vehicle at the initial time.
By using a polynomial model of order 5 for modeling the longitudinal trajectory of the self-vehicle and set constraints, it can be demonstrated that the end time thus obtained is the best compromise between dynamic comfort and physiological comfort. For example, when the speed of the own vehicle is 130km/h, the speed of the traveling preceding vehicle is 110km/h, the distance deviation at the initial time is 100 meters, and the inter-vehicle time is 2 seconds, the end time is equal to about 11 seconds.
Advantageously, the value of the classification parameter k is two-thirds and follows the relationBetween the values kMax of (c).
The smaller the classification parameter k, the smaller the manipulation time and thus the faster the manipulation. When the classification parameter k is less than two-thirds, the target acceleration is very dynamic, there is a strong acceleration followed by a strong deceleration. This is an uncomfortable "bang bang" type of maneuver. Based on a classification parameter k close to two thirds, there is no (positive) acceleration to be decelerated next. When the classification parameter k is close to two thirds, the maneuver is still fast, but the maneuver is considered anxiety by the driver. In fact, the most intense negative acceleration is reached towards the end of the maneuver. When the classification parameter is close to 1, the manipulation is considered to be psychologically pleasant. Using k close to kMax (typically, about 1.5), the change in the target acceleration trajectory is minimized, while the inter-vehicle distance temporarily becomes close to 0.
Advantageously, the result of the determination of the classification parameter k is a digital value proportional to a predetermined inter-vehicle time.
In the different modes of operation described above, the different modes of docking are implemented by using a limited number of parameters. Calibration or adjustment of the speed regulator is simplified. A value of the classification parameter k close to or slightly less than 1 determines the set acceleration trajectory that results in a quick docking. A value of said classification parameter k close to or slightly greater than 1 determines a set acceleration trajectory that results in a psychologically comfortable docking. The value of the classification parameter k between 1 and kmax determines the set acceleration trajectory that results in a normal abutment. The value of the classification parameter k close to kMax but less than kMax determines the set acceleration trajectory that results in a dynamic comfortable docking.
A second aspect of the invention relates to an apparatus comprising a memory associated with at least one processor configured to implement the method according to the first aspect of the invention.
The invention also relates to a vehicle comprising said device.
The invention also relates to a computer program comprising instructions adapted to perform the steps of the method according to the first aspect of the invention when said program is executed by at least one processor.
Drawings
Other features and advantages of the present invention will become more apparent upon reading the following detailed description of non-limiting embodiments of the invention and the accompanying drawings, in which:
figure 1 schematically shows an apparatus according to a specific embodiment of the invention.
Fig. 2 schematically shows a determination method for determining a set acceleration trajectory according to a specific embodiment of the invention.
Detailed Description
In the following, the invention is described in non-limiting application in the context of autonomous motor vehicles that travel on roads or traffic lanes. Other applications are possible, such as robots in storage warehouses or motorcycles on rural roads.
Fig. 1 shows an example of a device 101 included in a vehicle, in a network ("cloud"), or in a server. The device 101 may be used as a centralized device responsible for at least some of the steps of the method described below with reference to fig. 2. In an embodiment, the device corresponds to an autonomous driving computer.
In the present invention, the device 101 is included in the vehicle.
The device 101 may take the form of a box comprising printed circuitry, any type of computer or mobile phone ("smart phone").
The apparatus 101 comprises a random access memory 102 for storing instructions for implementing at least one step of the method as described above by the processor 103. The device also includes a mass memory 104 for storing data to be retained after implementation of the method.
The device 101 may also include a Digital Signal Processor (DSP) 105. The DSP105 receives data to shape, demodulate and amplify the data in a manner known per se.
The device 101 further comprises an input interface 106 for receiving data implemented by the method according to the invention and an output interface 107 for transmitting data implemented by the method according to the invention.
Fig. 2 schematically illustrates a determination method for determining a set acceleration trajectory of an adaptive vehicle speed regulator of an autonomous vehicle, referred to as a host vehicle, to dock with a vehicle traveling in front of the host vehicle, referred to as a preceding vehicle, according to a specific embodiment of the present invention.
Step 201 (Detk) is a determination step for determining a classification parameter k, which characterizes the docking behavior.
Different docking behaviors are possible. For example, one of the driving modes is a rapid docking, which is characterized by strong longitudinal vehicle dynamics (i.e. strong acceleration and deceleration to adapt the inter-vehicle distance as quickly as possible). Another driving mode is a slow and dynamically comfortable docking that is characterized by weak longitudinal vehicle dynamics. The intermediate driving mode is a psychologically comfortable dock characterized by a more intense longitudinal vehicle dynamics at the beginning of the maneuver at an initial time than at the end of the maneuver at an end time. In the last case, the driver perceives good detection of the target vehicle at the beginning of the maneuver.
The determination of the classification parameter k may be implemented by different methods. In an embodiment, in case there is a selection knob for selecting economy/normal/sport mode (e.g. it is present for classifying suspension comfort or for classifying an automatic gearbox or manually steering a gear change of a gearbox), a specific value is assigned to said classification parameter k according to the selection of said knob. For example, when a motion mode (or normal mode and economy mode, respectively) is selected, the classification parameter has a value of 1 (or 1.1, 1.3, respectively).
In another mode of operation, when the driver inputs the inter-vehicle time on the touch screen of the dashboard, then a specific value is assigned to the classification parameter k. For example, when the entered inter-vehicle time is 1 second (or 1.5 seconds and 2 seconds, respectively), then the classification parameter has a value of 1 (or 1.1 and 1.3, respectively).
Step 202 (Det Veh, t 0) is a detection step for detecting the traveling preceding vehicle that is on the same traffic lane as the own vehicle, the detection determining an initial time.
Thanks to the existing sensors present with the speed regulation system, the device is able to acquire the longitudinal dynamics of the own vehicle and of the preceding vehicle of travel, in particular when the system comprises ACC functionality. In particular, the device is able to receive detection information of the detection of a new vehicle (which travels in front of the own vehicle) on the same traffic lane as the own vehicle. The detected new vehicle becomes the traveling preceding vehicle. For example, these sensors are cameras, radar, lidar or any other device based on the processing of light waves, electromagnetic waves or acoustic information.
The initial time is determined at the moment a new vehicle is detected. In fact, the docking maneuver continues during time tf. Defining and determining a set acceleration trajectory requires defining a set acceleration at each time instant between the initial time and time tf.
In one mode of operation, the time tf has a predetermined value. The value may be different depending on the speed of the own vehicle. In another mode of operation, the time tf is determined based on longitudinal dynamics of the own vehicle and the traveling preceding vehicle.
Step 203 (Det Vc) is a determining step for determining a target speed Vc that characterizes a speed that the own vehicle needs to have at a predetermined end time tf when the own vehicle has docked the traveling preceding vehicle.
In a preferred mode of operation, the target speed is equal to the speed of the traveling preceding vehicle measured at the initial time. The speed of the traveling preceding vehicle is assumed to be substantially constant during a docking maneuver that lasts only a few seconds.
In another mode of operation, the target speed is derived based on measurements of speed and acceleration of the traveling preceding vehicle. In another mode of operation, step 203 is updated to cause an update of subsequent steps when needed.
Step 204 (Det D) is a determining step for determining a distance deviation D between the measured inter-vehicle distance D0 and a target inter-vehicle distance Dc, the target inter-vehicle distance being determined based on the target speed and a predetermined target inter-vehicle time.
In one mode of operation, the device receives measurement information of the inter-vehicle distance D0 at the initial time. The target inter-vehicle distance is determined based on the target inter-vehicle time and the target speed. The target inter-vehicle time is determined based on a prescribed default value (e.g. 2 seconds) or based on an input or based on an indirect selection performed by the driver on a human-machine interface of the vehicle, e.g. for selecting the classification parameter k.
In one mode of operation, the method further comprises a determining step for determining a speed deviation V between a measured speed V0 of the own vehicle and a target speed Vc and for equation-basedAnd a determining step of determining the end time tf. The end time tf and thus the docking maneuver duration are thus simply calculated and determined based on the inter-vehicle distance measured at the initial time, the speed of the own vehicle at the initial time, the target inter-vehicle distance and the target speed. Constraints in terms of limiting position, limiting speed, limiting acceleration, limiting jerk (jerk) are set, the equation being obtained by a polynomial model of order 5 based on the longitudinal trajectory of the vehicle.
Step 205 (Deta (t)) is a determination step for determining a set acceleration trajectory based on the distance deviation D, the end time tf, and the classification parameter k.
There are a number of possibilities for determining the acceleration trajectory. In one mode of operation, the acceleration is, for example, constant and proportional toProportional to the ratio. However, this may thus cause uncontrolled transition of the longitudinal dynamic behaviour of the self-vehicle at the beginning of the docking and at the end of the docking.
In a preferred mode of operation, the determination of the set acceleration trajectory a (t) is obtained based on the following equation:
where t is time and, at the initial time, t=0. In the last mode of operation, it can be demonstrated that the (acceleration, jerk) constraints are controlled and that, by only a change in the value k, a classified different docking behavior is possible. For example, the classification parameter k is between two thirds andand a value in between. Thus, when the classification parameter k is close to two thirds, the maneuver is rapid, but the maneuver is considered anxiety by the driver. In fact, the most intense negative acceleration is reached towards the end of the maneuver. When the classification parameter is close to 1, the manipulation is considered to be psychologically pleasant. For k=1.1, a normal/standard docking is obtained. For k=1.3, a slow and dynamically comfortable docking is obtained.
In another mode of operation, the result of the determination of the classification parameter k is a digital value proportional to a predetermined inter-vehicle time. The smaller the inter-vehicle time, the smaller the value of the classification parameter. Conversely, the greater the inter-vehicle time, the greater the value of the classification parameter. Thus, when the duration of the set acceleration trajectory depends on the end time tf, it is only necessary to multiply the time tf by the classification parameter k to accelerate or decelerate the operation.
The invention is not limited to the embodiments described above as examples; the invention extends to other variants.
Equations and calculations are also detailed. The invention is not limited to these equations and forms of computation, but extends to any type of other equivalent mathematical form.