CN114802302B - A processing method for trajectory planning based on multi-target predicted trajectories - Google Patents

Abstract

The embodiment of the present invention relates to a processing method for trajectory planning based on multi-target predicted trajectories, the method comprising: a planning module receives a multi-target predicted trajectory set output by an upstream prediction module at any time t ₀ ; any possible predicted trajectories of all targets in a future period T are combined and each combination is regarded as a possible future scene to obtain a scene trajectory combination f _x corresponding to multiple future scenes; the optimal driving trajectory of the vehicle in the corresponding scene x in the future period T is planned; the trajectory rationality check is performed; if the check is successful, the planned trajectory set is saved. Through the present invention, the optimal planned trajectory in any possible scene in the future period can be saved as the trajectory data to be checked, which can not only avoid the problem of only depositing one planned trajectory in the conventional processing method, but also provide a complete query basis for real-time trajectory query, thereby achieving the purpose of improving vehicle safety.

Description

Processing method for track planning based on multi-target predicted track

Technical Field

The invention relates to the technical field of data processing, in particular to a processing method for track planning based on multi-target predicted tracks.

Background

The upstream and downstream modules of the vehicle automatic driving system planning module are respectively as follows: a prediction module and a control module; the traditional prediction module predicts a plurality of possible motion tracks in future time periods of each target based on target (obstacle) track information acquired by a perception sensor, and sends a multi-target prediction track set obtained by prediction to the planning module; the traditional planning module carries out maximum probability estimation on various possible multi-target motion scenes in a future period based on the multi-target prediction track set, and outputs a unique track planned based on the maximum probability scene to the downstream control module; the traditional control module takes the unique track as a reference track and tracks the reference track, and the actual motion track is maximally close to the reference track by driving control (such as steering wheel control, accelerator/brake control and the like) in the track tracking process, namely, the real-time motion state of the bicycle is controlled. In the traditional processing mode, the vehicle automatic driving system only deposits one planning track in one working period, and if the target track mutation is encountered, that is, the multi-target actual motion scene is not consistent with the estimated maximum probability scene, the inquiry and the switching of the planning track can not be carried out based on the acquired multi-target actual motion track set. This clearly brings a major safety hazard to the driving control of the vehicle.

Disclosure of Invention

The invention aims at overcoming the defects of the prior art, and provides a processing method, electronic equipment and a computer readable storage medium for track planning based on multi-target predicted tracks, wherein a planning module firstly confirms scene track combinations of all possible multi-target motion scenes in a future period based on a multi-target predicted track set sent by an upstream prediction module, then plans optimal running tracks of a vehicle in each scene in the future period according to each scene track combination, performs track rationality check on a plurality of obtained optimal planning tracks, and forms all the optimal planning tracks into a corresponding planning track set and stores the planning track set after the check is successful. By the method and the device, the optimal planned track in any possible scene in the future period can be stored as track data to be checked, so that the problem that only one planned track is deposited in a conventional processing mode can be avoided, and a complete query basis can be provided for real-time track query, thereby achieving the aim of improving the safety guarantee of the vehicle.

To achieve the above object, a first aspect of an embodiment of the present invention provides a processing method for track planning based on a multi-target predicted track, where the method includes:

The planning module receives the multi-target prediction track set output by the upstream prediction module at any time t ₀; the predicted track period of the multi-target predicted track set is a future period T: [ t ₀,t₀+L],t₀ is the period start time, t ₀ +L is the period end time, and L is the period duration; the multi-target predicted track set comprises a plurality of first target track groups R; the first target track group R comprises a plurality of first target tracks p _i,j, i is a target index, j is a track index of a target i, i is more than or equal to 1 and less than or equal to n, j is more than or equal to 1 and less than or equal to m _i, n is a target number, and m _i is a track number of the target i;

Combining any one possible predicted track of all targets in the future period T according to the multi-target predicted track set and regarding each combination as a possible future scene, thereby obtaining scene track combinations f _x respectively corresponding to a plurality of future scenes; x is scene index, x is more than or equal to 1 and less than or equal to w, and w is scene number; the scene track combination f _x is composed of the first target tracks p _i,j of the target number n, and the target index i of each first target track p _i,j in the combination is not repeated;

planning the optimal running track of the own vehicle in the future period T in the corresponding scene x according to each scene track combination f _x to obtain a corresponding optimal planning track d _x;

performing track rationality check on all the optimal planning tracks d _x;

and if the track rationality check is successful, storing a planning track set consisting of all the optimal planning tracks d _x.

Preferably, each of the first target trajectories p _i,j is a possible motion trajectory of the corresponding target i in the future period T; a section of overlapping track is arranged between any two first target tracks p _i,j in each first target track group R from the initial position.

Preferably, the combining any one possible predicted track of all targets in the future period T according to the multi-target predicted track set and treating each combination as a possible future scene, so as to obtain a scene track combination f _x corresponding to a plurality of future scenes respectively, which specifically includes:

optionally, one of the first target track groups p _i,j in each of the first target track groups R of the multi-target track set is combined to form a corresponding scene track combination f _x;

Wherein each of the scene track combinations f _x corresponds to a possible future scene x in the future period T, respectively; between any two scene track combinations f _x, the combination relationship of the first target track p _i,j in each combination is not repeated; number of scenes As a combined function.

Preferably, the planning, according to each of the scene track combinations f _x, the optimal driving track of the own vehicle in the future period T in the corresponding scene x to obtain a corresponding optimal planned track d _x specifically includes:

Acquiring a starting reference position and a target reference position of the self-vehicle in the future period T;

Under the condition that collision with any one of the first target tracks p _i,j of the current scene track combination f _x is not met, planning a running track from the initial reference position to the target reference position of the own vehicle in the future period T, and obtaining one or more planned tracks d _x,a; a is a planning track index of a scene x, and a is more than or equal to 1;

counting the number of the planned tracks d _x,a to generate a corresponding first number;

Identifying the first number; if the first number is 1, outputting the unique planning track d _x,a as the optimal planning track d _x corresponding to the current scene x; if the first number is greater than 1, calculating the track cost of each planned track d _x,a according to a preset track cost calculation principle to generate corresponding first cost data, and outputting the planned track d _x,a corresponding to the first cost data with the minimum value as the optimal planned track d _x corresponding to the current scene x.

Preferably, the track rationality checking for all the optimal planned tracks d _x specifically includes:

Performing trace motion rationality check on all the optimal planning traces d _x; if the track movement direction check is successful, performing track branch time rationality check on any two optimal planning tracks d _x; and if the trace branching time check is successful, the trace rationality check is successful.

Further, the performing the trace motion rationality check on all the optimal planned traces d _x specifically includes:

On each optimal planning track d _x, the track point with the previous time and the track point with the next time in the adjacent two track points are marked as the last track point, the relative displacement of the next track point relative to the last track point is calculated, and if all the relative displacements of the current optimal planning track d _x are greater than 0, the corresponding checking state is set as the passing state; and if all the inspection states corresponding to the optimal planning tracks d _x are all passing states, the track movement rationality inspection is successful.

Further, the performing a trace branching time rationality check on any two of the optimal planned traces d _x specifically includes:

Optionally, the two optimal planned tracks d _x are respectively marked as corresponding first tracks and second tracks; the scene track combination f _x corresponding to the first track and the second track is respectively marked as a corresponding first combination and a corresponding second combination;

In the first and second combinations, confirming the coincident track of the two first target tracks p _i,j corresponding to each target i from the starting position, and taking the confirmed coincident track ending time as the corresponding first target track branching time t _1,i, so as to obtain n first target track branching times t _1,i; and selecting the first target track branching time t _1,i with the earliest time from the first target track branching time t _1,i as a first branching time t ₁ corresponding to the first and second combinations;

Confirming the coincident tracks of the first track and the second track from the initial position, and taking the confirmed coincident track ending time as a corresponding second branch time t ₂;

Calculating a time difference Δt=t ₂-t₁ between the second branch time t ₂ and the first branch time t ₁; identifying whether the time difference delta t meets a preset reasonable time difference range, and if the time difference delta t meets the reasonable time difference range, successfully checking the rationality of the track branching time; the reasonable time difference range is a time difference range greater than 0 and less than or equal to a specified threshold δ.

A second aspect of an embodiment of the present invention provides an electronic device, including: memory, processor, and transceiver;

The processor is configured to couple to the memory, and read and execute the instructions in the memory, so as to implement the method steps described in the first aspect;

The transceiver is coupled to the processor and is controlled by the processor to transmit and receive messages.

A third aspect of the embodiments of the present invention provides a computer-readable storage medium storing computer instructions that, when executed by a computer, cause the computer to perform the method of the first aspect described above.

The embodiment of the invention provides a processing method, electronic equipment and a computer readable storage medium for track planning based on multi-target predicted tracks, wherein a planning module firstly confirms scene track combinations of all possible multi-target motion scenes in a future period based on a multi-target predicted track set sent by an upstream prediction module, then plans optimal running tracks of a vehicle in each scene in the future period according to each scene track combination, performs track rationality check on a plurality of obtained optimal planning tracks, and forms and stores all the optimal planning tracks into corresponding planning track sets after the check is successful. By the method and the device, the optimal planned track in any possible scene in the future period can be stored as track data to be checked, so that the problem that only one planned track is deposited in a conventional processing mode is avoided, a complete query basis is provided for real-time track query, and vehicle safety guarantee is improved.

Drawings

Fig. 1 is a schematic diagram of a processing method for track planning based on a multi-target predicted track according to a first embodiment of the present invention;

FIG. 2 is a schematic diagram of a planned trajectory according to a first embodiment of the present invention;

fig. 3 is a schematic structural diagram of an electronic device according to a second embodiment of the present invention.

Detailed Description

In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention will be described in further detail below with reference to the accompanying drawings, and it is apparent that the described embodiments are only some embodiments of the present invention, not all embodiments. 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.

An embodiment of the present invention provides a processing method for performing track planning based on a multi-target predicted track, as shown in fig. 1, which is a schematic diagram of a processing method for performing track planning based on a multi-target predicted track according to an embodiment of the present invention, the method mainly includes the following steps:

step 1, a planning module receives a multi-target prediction track set output by an upstream prediction module at any time t ₀;

The predicted track period of the multi-target predicted track set is a future period T: [ t ₀,t₀+L],t₀ is the period start time, t ₀ +L is the period end time, and L is the period duration; the multi-target predicted track set comprises a plurality of first target track groups R; the first target track group R comprises a plurality of first target tracks p _i,j, i is a target index, j is a track index of a target i, i is more than or equal to 1 and less than or equal to n, j is more than or equal to 1 and less than or equal to m _i, n is a target number, and m _i is a track number of the target i; each first target locus p _i,j is a possible motion locus of the corresponding target i in the future period T; a section of coincident track is arranged between any two first target tracks p _i,j in each first target track group R from the starting position.

Here, the multi-target predicted trajectory set is a complete set of a plurality of possible motion trajectories of all obstacles (targets) in the traffic environment where the own vehicle is located in the future period T; each first target track group R corresponds to a target i, and each first target track p _i,j in the track group corresponds to a possible motion track of the target i; in the embodiment of the present invention, the time point when each target makes various possible track switching (abrupt change) in the future period T is after the starting time point T ₀, so that the head positions of the first target tracks p _i,j corresponding to the same target must have a section of track overlapped, so that any two first target tracks p _i,j in each first target track group R have a section of overlapped track from the starting position (the track position corresponding to the time T ₀).

For example, two obstacle targets around the host vehicle are vehicles 1,2, and the upstream prediction module predicts that there are 2 possible movements (straight, left turn) of the vehicle 1 at time T _a in the future period T, whereas there are only 1 possible movements (straight) of the vehicle 2 throughout the future period T; then, the number of targets n=2, the number of tracks m _i＝1 =2 of target i=1, and the number of tracks m _i＝2 =1 of target i=2, i.e. the multi-target predicted track set should include 2 first target track groups R1, R2, R1 having 2 first target tracks p _1,1、p_1,2, R2 having 1 first target track p _2,1, the tracks of the first target tracks p _1,1、p_1,2 in the first target track group R1 being coincident in the period t ₀,t_a.

Step 2, according to the multi-target predicted track set, combining any one possible predicted track of all targets in the future period T and regarding each combination as a possible future scene, thereby obtaining scene track combinations f _x corresponding to a plurality of future scenes respectively;

The method specifically comprises the following steps: combining the first target tracks p _i,j from each first target track group R of the multi-target track set to form a corresponding scene track combination f _x;

wherein x is scene index, x is more than or equal to 1 and less than or equal to w, and w is scene number; the scene track combination f _x consists of a first target track p _i,j with the target number n, and the target index i of each first target track p _i,j in the combination is not repeated; each scene track combination f _x corresponds to one possible future scene x in the future period T, respectively; the combination relation of the first target track p _i,j in each combination is not repeated between any two scene track combinations f _x; number of scenes As a combined function, i.e

For example, the multi-target predicted track set for the known future period T includes 2 first target track groups R1, R2, R1 having 2 first target tracks p _1,1、p_1,2, R2 having 1 first target track p _2,1;

Then, the number of scenes can be calculated by combining the first target track p _i,j selected from the first target track groups R of the multi-target track set The corresponding two scene track combinations f _x＝1、f_x＝2 are: f ₁(p_1,1,p_1,2)、f₂(p_1,2,p_1,2).

Step 3, planning the optimal running track of the own vehicle in the corresponding scene x in the future period T according to each scene track combination f _x to obtain a corresponding optimal planning track d _x;

The method specifically comprises the following steps: step 31, acquiring a starting reference position and a target reference position of the own vehicle in a future period T;

here, the initial reference position and the target reference position of the own vehicle in the future period T are actually the current position coordinates of the own vehicle and the position coordinates of the traveling destination of the own vehicle; the two pieces of position information can be acquired from a positioning module and a map module of the automatic driving system of the vehicle;

Step 32, under the condition that any one of the first target tracks p _i,j of the current scene track combination f _x is not collided, planning a running track from a starting reference position to a target reference position of the own vehicle in a future period T, and obtaining one or more planned tracks d _x,a;

wherein a is the planning track index of the scene x, and a is not less than 1;

The planning algorithm for planning the driving track is similar to the path planning and motion state planning algorithm used by the conventional planning module, and is not further described herein; it should be noted that, in the current step 32, a corresponding vehicle track is planned for each multi-target motion scene that may occur in the future, and each scene track combination f _x corresponds to one scene that may occur in the future and includes a possible motion track of each of all the obstacle targets (e.g., vehicles 1 and 2) in the corresponding scene; while a plurality of planned trajectories d _x,a may occur during trajectory planning;

Step 33, counting the number of the planned tracks d _x,a to generate a corresponding first number;

For example, as shown in the schematic diagram of the planned track provided in fig. 2 for the first embodiment of the present invention, it is known that two obstacle targets are vehicles 1 and 2 around the vehicle, and that there are 2 possible movements (straight and left turn) of the vehicle 1 at T _a in the future period T, and that there are only 1 possible movements (straight) of the vehicle 2 during the whole future period T, the multi-target predicted track set of the future period T includes 2 first target track groups R1 and R2, and that the R1 has 2 first target tracks p _1,1、p_1,2 and the R2 has 1 first target track p _2,1, so as to obtain 2 scene track combinations f _x＝1、f_x＝2 as follows: f ₁(p_1,1,p_1,2)、f₂(p_1,2,p_1,2); setting that 2 planning tracks are d _1,1、d_1,2 respectively based on f ₁(p_1,1,p_1,2) planning and 1 planning track is d _2,1 based on f ₂(p_1,2,p_1,2) planning;

then, when the current scene track combination f _x is f ₁(p_1,1,p_1,2), the number of corresponding planned tracks d _x,a, that is, the first number is 2; when the current scene track combination f _x is f ₂(p_1,2,p_1,2), the number of the corresponding planned tracks d _x,a, that is, the first number is 1;

Step 34, identifying a first number; if the first number is 1, outputting the unique planning track d _x,a as an optimal planning track d _x corresponding to the current scene x; if the first number is greater than 1, calculating the track cost of each planned track d _x,a according to a preset track cost calculation principle to generate corresponding first cost data, and outputting the planned track d _x,a corresponding to the first cost data with the minimum value as an optimal planned track d _x corresponding to the current scene x.

Here, in the embodiment of the invention, only one optimal planning track is output under each scene x, and the optimal track screening principle is a preset track cost calculation principle, and the principle has various setting standards and implementation modes; one of the rules may be that the overall displacement distance of the track is the minimum value, namely, the corresponding track cost is calculated based on the overall displacement distance of each planned track d _x,a, the larger the displacement is limited, the larger the cost is, and finally, the planned track d _x,a corresponding to the first generation price data with the minimum value, namely, the planned track d _x,a with the minimum overall displacement distance of the track is output as an optimal planned track d _x; the other principle is that the whole track energy consumption is taken as the minimum value principle, namely, the corresponding track cost is calculated based on the whole track energy consumption of each planned track d _x,a, the larger the energy consumption is, the larger the cost is, and finally, the planned track d _x,a corresponding to the first generation price data with the minimum value, namely, the planned track d _x,a with the minimum whole track energy consumption is output as the optimal planned track d _x.

Taking fig. 2 as an example, when the current scene track combination f _x is f ₁(p_1,1,p_1,2), 2 corresponding planned tracks d _1,1、d_1,2 are obtained, and the track cost of d _1,1、d_1,2 is calculated by taking the track overall energy as the minimum value principle as the track cost calculation principle to obtain first cost data corresponding to d _1,1, wherein the first cost data corresponding to d _1,1 is smaller than the first cost data corresponding to d _1,2, so d _1,1 is selected as the optimal planned track d ₁ corresponding to f ₁(p_1,1,p_1,2); when the current scene track combination f _x is f ₂(p_1,2,p_1,2), 1 planned track d _2,1 is obtained, and d _2,1 is directly taken as the optimal planned track d ₂ with f ₂(p_1,2,p_1,2).

Step 4, performing track rationality check on all the optimal planning tracks d _x;

The method specifically comprises the following steps: step 41, performing trace motion rationality check on all the optimal planning traces d _x;

The method specifically comprises the following steps: on each optimal planning track d _x, marking the previous track point and the next track point which are positioned at the front and the back of two adjacent track points, calculating the relative displacement of the next track point relative to the previous track point, and setting the corresponding checking state as a passing state if all the relative displacements of the current optimal planning track d _x are larger than 0; if all the inspection states corresponding to all the optimal planning tracks d _x are passing states, the track movement rationality inspection is successful;

Here, the optimal planned trajectory d _x may be understood as being formed by arranging a plurality of trajectory points in chronological order; all relative displacements of the current optimal planning track d _x are larger than 0, namely the position of the own vehicle is always changed from the first track point to the last track point on the current optimal planning track d _x, namely the own vehicle is always in a motion state rather than a stop state, namely the current optimal planning track d _x is a normal planning track which can continuously approach from the initial position of the own vehicle to the target reference position, so that the corresponding checking state is set to be a passing state at the moment; the inspection states corresponding to all the optimal planned trajectories d _x are all passing states, so that all the optimal planned trajectories d _x are normal planned trajectories, and the trajectory movement rationality inspection is successful;

Step 42, checking the track movement direction successfully, and checking the track branching time rationality of any two optimal planning tracks d _x;

the method specifically comprises the following steps: step 421, selecting two optimal planning tracks d _x to be respectively marked as corresponding first tracks and second tracks; and the scene track combination f _x corresponding to the first track and the second track is respectively marked as a corresponding first combination and a corresponding second combination;

Taking fig. 2 as an example, only 2 optimal planned trajectories d ₁、d₂ are shown as the first and second trajectories d ₁、d₂; the corresponding first and second combinations are f ₁(p_1,1,p_1,2)、f₂(p_1,2,p_1,2);

step 422, in the first and second combinations, confirming the overlapping track of the two first target tracks p _i,j corresponding to each target i from the starting position, and taking the confirmed overlapping track ending time as the corresponding first target track branching time t _1,i, so as to obtain n first target track branching times t _1,i; and selecting a first target track branching time t _1,i with the earliest time from the first target track branching times as a first branching time t ₁ corresponding to the first and second combinations;

here, it should be noted that if two first target trajectories p _i,j of the target i in the first and second combinations are the same, the corresponding first target trajectory branching time t _1,i should be the trajectory ending time (t ₀ +l);

Taking fig. 2 as an example, there are 2 targets in total, namely vehicles 1 and 2, and the first and second combinations are f ₁(p_1,1,p_1,2)、f₂(p_1,2,p_1,2); the first target track branch time t _1,i＝1＝t_a of the vehicle 1 is when the 2 first target tracks corresponding to the vehicle 1 in the first and second combinations are p _1,1、p_1,2, the p _1,1、p_1,2 previous track of the vehicle 1 is overlapped and separated from the track point time t _a; the first target track branch time t _1,i＝2 of the vehicle 2 is the track end time (t ₀ +l), if the first target tracks corresponding to the vehicle 2 in the first and second combinations are p _2,1; since t _1,i＝1 is earlier in the resulting 2 first target track branch times t _1,i＝1、t_1,i＝2, t _1,i＝1＝t_a is taken as the first branch time t ₁;

Step 423, confirming the coincident tracks of the first track and the second track from the initial position, and taking the confirmed coincident track ending time as the corresponding second branch time t ₂;

Taking fig. 2 as an example, if only 2 optimal planned trajectories d ₁、d₂ are d ₁、d₂, the first and second trajectories d ₁、d₂ are overlapped and separated from the trajectory point with the trajectory point time t _b, and then the second branching time t ₂＝t_b is known;

Step 424, calculating a time difference Δt=t ₂-t₁ between the second branch time t ₂ and the first branch time t ₁; identifying whether the time difference delta t meets a preset reasonable time difference range, and if the time difference delta t meets the reasonable time difference range, successfully checking the rationality of the track branch time;

Wherein the reasonable time difference range is a time difference range greater than 0 and less than or equal to a specified threshold delta;

Here, in the embodiment of the present invention, the planned track branching time and the predicted track branching time are constrained by t ₁<t₂≤(t₁ +δ), that is, the planned track branching time is required to be later than the predicted track branching time; designating a threshold delta as a system delay parameter, and setting by default according to the overall perception delay of the vehicle perception sensor and the perception module;

Step 43, the trace branching time check is successful and the trace rationality check is successful.

And 5, if the track rationality check is successful, storing a planning track set consisting of all the optimal planning tracks d _x.

Fig. 3 is a schematic structural diagram of an electronic device according to a second embodiment of the present invention. The electronic device may be the aforementioned terminal device or server, or may be a terminal device or server connected to the aforementioned terminal device or server for implementing the method of the embodiment of the present invention. As shown in fig. 3, the electronic device may include: a processor 301 (e.g., a CPU), a memory 302, a transceiver 303; the transceiver 303 is coupled to the processor 301, and the processor 301 controls the transceiving actions of the transceiver 303. The memory 302 may store various instructions for performing the various processing functions and implementing the processing steps described in the method embodiments previously described. Preferably, the electronic device according to the embodiment of the present invention further includes: a power supply 304, a system bus 305, and a communication port 306. The system bus 305 is used to implement communication connections between the elements. The communication port 306 is used for connection communication between the electronic device and other peripheral devices.

The system bus 305 referred to in fig. 3 may be a peripheral component interconnect standard (PERIPHERAL COMPONENT INTERCONNECT, PCI) bus, or an extended industry standard architecture (Extended Industry Standard Architecture, EISA) bus, or the like. The system bus may be classified into an address bus, a data bus, a control bus, and the like. For ease of illustration, only one thick line is shown in fig. 3, but not only one bus or one type of bus. The communication interface is used to enable communication between the database access apparatus and other devices (e.g., clients, read-write libraries, and read-only libraries). The Memory may include random access Memory (Random Access Memory, RAM) and may also include Non-Volatile Memory (Non-Volatile Memory), such as at least one disk Memory.

The processor may be a general-purpose processor, including a central processing unit (Central Processing Unit, CPU), a network processor (Network Processor, NP), a graphics processor (Graphics Processing Unit, GPU), etc.; but may also be a digital signal Processor (DIGITAL SIGNAL Processor, DSP), application SPECIFIC INTEGRATED Circuit (ASIC), field programmable gate array (Field Programmable GATE ARRAY, FPGA) or other programmable logic device, discrete gate or transistor logic device, discrete hardware components.

It should be noted that, the embodiments of the present invention also provide a computer readable storage medium, where instructions are stored, when the computer readable storage medium runs on a computer, to cause the computer to perform the method and the process provided in the above embodiments.

The embodiment of the invention also provides a chip for running the instructions, and the chip is used for executing the processing steps described in the embodiment of the method.

The embodiment of the invention provides a processing method, electronic equipment and a computer readable storage medium for track planning based on multi-target predicted tracks, wherein a planning module firstly confirms scene track combinations of all possible multi-target motion scenes in a future period based on a multi-target predicted track set sent by an upstream prediction module, then plans optimal running tracks of a vehicle in each scene in the future period according to each scene track combination, performs track rationality check on a plurality of obtained optimal planning tracks, and forms and stores all the optimal planning tracks into corresponding planning track sets after the check is successful. By the method and the device, the optimal planned track in any possible scene in the future period can be stored as track data to be checked, so that the problem that only one planned track is deposited in a conventional processing mode is avoided, a complete query basis is provided for real-time track query, and vehicle safety guarantee is improved.

Those of skill would further appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both, and that the various illustrative elements and steps are described above generally in terms of function in order to clearly illustrate the interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.

The steps of a method or algorithm described in connection with the embodiments disclosed herein may be embodied in hardware, in a software module executed by a processor, or in a combination of the two. The software modules may be disposed in Random Access Memory (RAM), memory, read Only Memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.

The foregoing description of the embodiments has been provided for the purpose of illustrating the general principles of the invention, and is not meant to limit the scope of the invention, but to limit the invention to the particular embodiments, and any modifications, equivalents, improvements, etc. that fall within the spirit and principles of the invention are intended to be included within the scope of the invention.

Claims (9)

1. A method for processing trajectory planning based on a multi-target predicted trajectory, the method comprising:

The planning module receives the multi-target prediction track set output by the upstream prediction module at any time t ₀; the predicted track period of the multi-target predicted track set is a future period T: [ t ₀,t₀+L],t₀ is the period start time, t ₀ +L is the period end time, and L is the period duration; the multi-target predicted track set comprises a plurality of first target track groups R; the first target track group R comprises a plurality of first target tracks p _i,j, i is a target index, j is a track index of a target i, i is more than or equal to 1 and less than or equal to n, j is more than or equal to 1 and less than or equal to m _i, n is a target number, and m _i is a track number of the target i;

Combining any one possible predicted track of all targets in the future period T according to the multi-target predicted track set and regarding each combination as a possible future scene, thereby obtaining scene track combinations f _x respectively corresponding to a plurality of future scenes; x is scene index, x is more than or equal to 1 and less than or equal to w, and w is scene number; the scene track combination f _x is composed of the first target tracks p _i,j of the target number n, and the target index i of each first target track p _i,j in the combination is not repeated;

planning the optimal running track of the own vehicle in the future period T in the corresponding scene x according to each scene track combination f _x to obtain a corresponding optimal planning track d _x;

performing track rationality check on all the optimal planning tracks d _x;

and if the track rationality check is successful, storing a planning track set consisting of all the optimal planning tracks d _x.

2. The method of claim 1, wherein the method further comprises the steps of,

Each of the first target trajectories p _i,j is a possible motion trajectory of the corresponding target i in the future period T; a section of overlapping track is arranged between any two first target tracks p _i,j in each first target track group R from the initial position.

3. The method according to claim 1, wherein the combining any one of the possible predicted trajectories of all the targets in the future period T according to the multi-target predicted trajectory set and treating each combination as one possible future scene, thereby obtaining scene trajectory combinations f _x corresponding to a plurality of future scenes, respectively, specifically includes:

optionally, one of the first target track groups p _i,j in each of the first target track groups R of the multi-target track set is combined to form a corresponding scene track combination f _x;

Wherein each of the scene track combinations f _x corresponds to a possible future scene x in the future period T, respectively; between any two scene track combinations f _x, the combination relationship of the first target track p _i,j in each combination is not repeated; number of scenes As a combined function.

4. The method for performing track planning based on multi-target predicted tracks according to claim 1, wherein the step of planning the optimal driving track of the own vehicle in the future period T in the corresponding scene x according to each scene track combination f _x to obtain a corresponding optimal planned track d _x specifically comprises:

Acquiring a starting reference position and a target reference position of the self-vehicle in the future period T;

Under the condition that collision with any one of the first target tracks p _i,j of the current scene track combination f _x is not met, planning a running track from the initial reference position to the target reference position of the own vehicle in the future period T, and obtaining one or more planned tracks d _x,a; a is a planning track index of a scene x, and a is more than or equal to 1;

counting the number of the planned tracks d _x,a to generate a corresponding first number;

Identifying the first number; if the first number is 1, outputting the unique planning track d _x,a as the optimal planning track d _x corresponding to the current scene x; if the first number is greater than 1, calculating the track cost of each planned track d _x,a according to a preset track cost calculation principle to generate corresponding first cost data, and outputting the planned track d _x,a corresponding to the first cost data with the minimum value as the optimal planned track d _x corresponding to the current scene x.

5. The method for performing track planning based on multi-objective predicted tracks according to claim 1, wherein the performing a track rationality check on all the optimal planned tracks d _x specifically comprises:

Performing trace motion rationality check on all the optimal planning traces d _x; if the track movement direction check is successful, performing track branch time rationality check on any two optimal planning tracks d _x; and if the trace branching time check is successful, the trace rationality check is successful.

6. The method for performing track planning based on multi-objective predicted track according to claim 5, wherein the performing a track motion rationality check on all the optimal planned tracks d _x specifically comprises:

On each optimal planning track d _x, the track point with the previous time and the track point with the next time in the adjacent two track points are marked as the last track point, the relative displacement of the next track point relative to the last track point is calculated, and if all the relative displacements of the current optimal planning track d _x are greater than 0, the corresponding checking state is set as the passing state; and if all the inspection states corresponding to the optimal planning tracks d _x are all passing states, the track movement rationality inspection is successful.

7. The method for performing track planning based on multi-objective predicted tracks according to claim 5, wherein the performing a track branching time rationality check on any two of the optimal planned tracks d _x specifically comprises:

Optionally, the two optimal planned tracks d _x are respectively marked as corresponding first tracks and second tracks; the scene track combination f _x corresponding to the first track and the second track is respectively marked as a corresponding first combination and a corresponding second combination;

In the first and second combinations, confirming the coincident track of the two first target tracks p _i,j corresponding to each target i from the starting position, and taking the confirmed coincident track ending time as the corresponding first target track branching time t _1,i, so as to obtain n first target track branching times t _1,i; and selecting the first target track branching time t _1,i with the earliest time from the first target track branching time t _1,i as a first branching time t ₁ corresponding to the first and second combinations;

Confirming the coincident tracks of the first track and the second track from the initial position, and taking the confirmed coincident track ending time as a corresponding second branch time t ₂;

Calculating a time difference Δt=t ₂-t₁ between the second branch time t ₂ and the first branch time t ₁; identifying whether the time difference delta t meets a preset reasonable time difference range, and if the time difference delta t meets the reasonable time difference range, successfully checking the rationality of the track branching time; the reasonable time difference range is a time difference range greater than 0 and less than or equal to a specified threshold δ.

8. An electronic device, comprising: memory, processor, and transceiver;

the processor being adapted to be coupled to the memory, read and execute the instructions in the memory to implement the method steps of any one of claims 1-7;

The transceiver is coupled to the processor and is controlled by the processor to transmit and receive messages.

9. A computer readable storage medium storing computer instructions which, when executed by a computer, cause the computer to perform the instructions of the method of any one of claims 1-7.