CN114633765A - Speed decision method and device based on probability grid graph and related products - Google Patents
Speed decision method and device based on probability grid graph and related products Download PDFInfo
- Publication number
- CN114633765A CN114633765A CN202210272793.7A CN202210272793A CN114633765A CN 114633765 A CN114633765 A CN 114633765A CN 202210272793 A CN202210272793 A CN 202210272793A CN 114633765 A CN114633765 A CN 114633765A
- Authority
- CN
- China
- Prior art keywords
- grid
- speed decision
- grid point
- probability
- path
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W60/00—Drive control systems specially adapted for autonomous road vehicles
- B60W60/001—Planning or execution of driving tasks
- B60W60/0011—Planning or execution of driving tasks involving control alternatives for a single driving scenario, e.g. planning several paths to avoid obstacles
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W2720/00—Output or target parameters relating to overall vehicle dynamics
- B60W2720/10—Longitudinal speed
Landscapes
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Human Computer Interaction (AREA)
- Transportation (AREA)
- Mechanical Engineering (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a speed decision method based on a probability grid diagram, a device thereof and a related product, wherein the method comprises the following steps: establishing an ST probability grid map in a freset coordinate system by taking a self-parking position as an origin; expanding a plurality of expanded grid points meeting preset conditions according to search step length by taking the grid point where the self-vehicle position is as a father node in the ST probability grid graph; taking each expansion grid point as a father node, and continuing to expand a plurality of expansion grid points meeting preset conditions according to the search step length; searching round by round until a preset termination condition is met to obtain a plurality of paths and corresponding speed decision results; for each path, calculating the total cost of the path according to the occupation probability of each grid point contained in the path and the relative distance between the path and the obstacle; determining a plurality of candidate speed decision results according to the total cost corresponding to the plurality of paths; and selecting a target speed decision result from the plurality of candidate speed decision results. The embodiment of the invention can better ensure the safety of the final speed decision result.
Description
Technical Field
The invention relates to the technical field of automatic driving, in particular to a speed decision method and a device thereof based on a probability grid graph and a related product.
Background
With the development of artificial intelligence technology, the automatic driving technology is more mature. The automatic driving technology can be briefly divided into perception, prediction, positioning, decision, planning and control. The decision planning is one of key parts of automatic driving, firstly needs to fuse multi-sensing information, then carries out task decision according to driving requirements, then plans a plurality of selectable safety paths between two points through some specific constraint conditions on the premise of avoiding possible obstacles, and selects an optimal path from the paths as a vehicle driving track. The decision technology can guarantee the driving safety of the unmanned vehicle and obey the traffic rules, and can also provide limiting information for smooth optimization of the path and the speed.
Frequent interaction with other traffic participants can exist in the driving process of the vehicle, such as scenes of front vehicle cut-in, pedestrian giving and intersection passing. For an automatic driving vehicle, the target speed needs to be adjusted in time to avoid conflict with other dynamic targets, speed planning can be performed in real time based on the surrounding environment, but the speed planning is difficult to effectively refer to historical information and accords with the driving habits of human beings. Therefore, speed regulation or behavior decision results can be made for specific obstacles or positions through speed decision combining with information such as history and traffic rules, and a clear planning target is provided for speed planning so as to improve the smoothness and the reasonability of vehicle performance.
In the related art, a commonly used speed decision method includes setting a virtual parking spot, a search algorithm based on barrier projection, a markov decision process, and the like. The method for setting the virtual parking point judges whether collision risks exist according to collision time of the vehicle and other obstacles, and if the risks exist, the virtual parking point is set or updated before the nearest collision point to achieve a line giving function. And projecting the sensing and predicting results of the obstacles in the speed planning space of the obstacles by a search algorithm based on obstacle projection, and designing a cost function to obtain a decision result for the conflict obstacles by adopting the search algorithm. And in the Markov decision process, according to the states of the own vehicle and the barrier, predicting the action in a period of time in the future and calculating the income, and taking the action with the maximum expected income as the current speed decision result.
The inventor finds that: some schemes in the prior art cannot guarantee the safety of the decision result, and further some schemes cannot guarantee the stability of the decision result.
Disclosure of Invention
The embodiment of the invention aims to solve at least one of the technical problems.
In a first aspect, an embodiment of the present invention provides a speed decision method based on a probability grid map, including: establishing an ST probability grid map in a frenet coordinate system by taking the self-parking position as an origin; expanding a plurality of expanded grid points meeting preset conditions according to search step length by taking the grid point where the self-parking position is as a father node in the ST probability grid graph; continuously expanding a plurality of expanded grid points which meet preset conditions according to the search step length by taking each expanded grid point as a father node; searching round by round until a preset termination condition is met to obtain a plurality of paths and corresponding speed decision results; for each path, calculating the total cost of the path according to the occupation probability of each grid point contained in the path and the relative distance between the path and an obstacle; determining a plurality of candidate speed decision results according to the total cost corresponding to the plurality of paths; and selecting a target speed decision result from the plurality of candidate speed decision results.
In a second aspect, an embodiment of the present invention provides a velocity decision apparatus based on a probability grid map, including: the probability grid map establishing module is used for establishing an ST probability grid map in a freset coordinate system by taking a self-parking position as an origin; the expansion module is used for expanding a plurality of expansion grid points meeting preset conditions according to search step length by taking the grid point where the self-parking position is located as a father node in the ST probability grid map; taking each expansion grid point as a father node, and continuing to expand a plurality of expansion grid points meeting preset conditions according to the search step length; searching round by round until a preset termination condition is met to obtain a plurality of paths and corresponding speed decision results; the total cost calculation module is used for calculating the total cost of each path according to the occupation probability of each grid point contained in the path and the relative distance between the path and an obstacle; a candidate determining module, configured to determine a plurality of candidate speed decision results according to the total costs corresponding to the plurality of paths; and a decision module for selecting a target speed decision result from the plurality of candidate speed decision results.
In a third aspect, an embodiment of the present invention provides an electronic device, including: at least one processor, and a memory communicatively coupled to the at least one processor, wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform any of the above-described probabilistic grid graph-based speed decision methods of the present invention.
In a fourth aspect, an embodiment of the present invention provides a mobile device, including a body and the electronic device according to any embodiment of the present invention mounted on the body.
In a fifth aspect, an embodiment of the present invention provides a storage medium, in which one or more programs including execution instructions are stored, where the execution instructions can be read and executed by an electronic device (including but not limited to a computer, a server, or a network device, etc.) to perform any one of the above-mentioned probabilistic grid graph-based speed decision methods of the present invention.
In a sixth aspect, an embodiment of the present invention further provides a computer program product, which includes a computer program stored on a storage medium, the computer program including program instructions, which when executed by a computer, cause the computer to perform any one of the above speed decision methods based on a probability grid map.
According to the method and the device, the ST probability grid map is established, and the occupation probability of the grid points in the ST probability grid map is used in the process of expanding and calculating the total cost, so that the speed decision result can better avoid the grid points with high occupation probability and keep away from the grids with the occupation probability as far as possible, the self-vehicle can always keep a safe relative distance with the moving obstacle, and the safety of the final speed decision result can be guaranteed.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
FIG. 1 is a flow diagram of one embodiment of a probabilistic grid graph-based speed decision method of the present invention;
FIG. 2 is a flow chart of another embodiment of a probabilistic grid graph-based speed decision method of the present invention;
FIG. 3 is a flow chart of yet another embodiment of a probabilistic grid graph-based speed decision method of the present invention;
FIG. 4 is a flow chart of yet another embodiment of a probabilistic grid graph-based speed decision method of the present invention;
FIG. 5 is a flow chart of yet another embodiment of a probabilistic grid graph-based speed decision method of the present invention;
FIG. 6 is a schematic diagram of a possible decision result in an ST probability grid diagram according to an embodiment of the present invention;
FIG. 7 is a flowchart of a feasible solution search based on a probability grid graph according to an embodiment of the present invention;
FIG. 8 is a block diagram of a velocity decision device based on a probability grid map according to an embodiment of the present invention;
fig. 9 is a schematic structural diagram of an embodiment of an electronic device according to the invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
It should be noted that, in the present application, the embodiments and features of the embodiments may be combined with each other without conflict.
The invention may be described in the general context of computer-executable instructions, such as program modules, being executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types. The invention may also be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media including memory storage devices.
As used in this disclosure, "module," "device," "system," and the like are intended to refer to a computer-related entity, either hardware, a combination of hardware and software, or software in execution. In particular, for example, an element may be, but is not limited to being, a process running on a processor, an object, an executable, a thread of execution, a program, and/or a computer. Also, an application or script running on a server, or a server, may be an element. One or more elements may be in a process and/or thread of execution and an element may be localized on one computer and/or distributed between two or more computers and may be operated by various computer-readable media. The elements may also communicate by way of local and/or remote processes based on a signal having one or more data packets, e.g., from a data packet interacting with another element in a local system, distributed system, and/or across a network in the internet with other systems by way of the signal.
Finally, it should also be noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
The embodiment of the invention provides a speed decision method based on a probability grid map, which can be applied to any mobile tool capable of realizing automatic driving, such as an automatic driving vehicle (a passenger vehicle, a bus, a small bus, a truck, an off-road vehicle, a sanitation vehicle, a sweeper, a ground cleaning vehicle, a dust collector and the like), a sweeping robot and the like, and the invention is not limited thereto.
Referring to fig. 1, a speed decision method based on a probability grid chart according to an embodiment of the present invention is shown.
As shown in fig. 1, in step 101, an ST probability grid map is established in a frient coordinate system with a self-parking location as an origin;
in step 102, expanding a plurality of expanded grid points meeting preset conditions according to search step length by taking the grid point where the self-parking position is located as a father node in the ST probability grid map; continuously expanding a plurality of expanded grid points which meet preset conditions according to the search step length by taking each expanded grid point as a father node; searching round by round until a preset termination condition is met to obtain a plurality of paths and corresponding speed decision results;
in step 103, for each path, calculating a total cost of the path according to an occupation probability of each grid point included in the path and a relative distance between the path and an obstacle;
in step 104, determining a plurality of candidate speed decision results according to the total costs corresponding to the plurality of paths;
in step 105, a target speed decision result is selected from the plurality of candidate speed decision results.
In this embodiment, for step 101, an ST probability grid map needs to be established in a frient coordinate system with a self-vehicle position as an origin, and in the ST probability grid map, a vertical coordinate S is a longitudinal distance of the obstacle after being projected relative to the reference path at a time point of an abscissa T.
In some embodiments, the aforementioned search step length may be set to 1s, for example, and the preset condition may be a series of conditions such that the own vehicle does not violate traffic regulations or perform risk driving while traveling, for example, the own vehicle does not exceed a speed limit of the current road segment, does not run a red light, and the like, which is not limited herein. The preset termination condition may be that a preset search step length is reached, a search time reaches a preset duration, and the like, and the application is not limited herein. Searching under the preset condition can reduce the number of expanded grid points while ensuring safe driving, and can limit the number of obtained paths, wherein each path corresponds to a speed decision result, the speed decision result can be, for example, overtaking obstacle a and obstacle B, letting go obstacle a and obstacle B, overtaking obstacle a and letting go obstacle B, etc., the number of obstacles can also be one or more, the speed decision result can also be in other possible combination modes, and the application is not limited herein.
Then, for step 103, for each path, the total cost of the path is calculated according to the occupation probability of each grid point included in the path and the relative distance between the path and the obstacle. Each grid point on each path may have an occupation probability occupied by one or more obstacles and a relative distance from each obstacle, and the total cost of the path may be calculated based on the occupation probability and the relative distance, for example, for grid point 1, the occupation probability occupied by obstacle a in one search step is 0.4, the occupation probability occupied by another obstacle B is 0, and similarly, the occupation probability of each grid point on the path occupied by an obstacle may be obtained, and the total cost may be calculated by a potential function based on the occupation probability and the relative distance, the potential function being, for example, a repulsive potential function or a gravitational potential function, which is not limited herein. By introducing the occupation probability and the relative distance in the speed decision process, the target speed decision result can avoid grid points with high occupation probability, keep away from the grid points with the occupation probability, and enable the self-vehicle to keep a safe distance with the obstacle all the time.
Then, for step 104, a plurality of candidate speed decision results are determined according to the total costs corresponding to the plurality of paths. Since each path corresponds to one candidate speed decision result, there are multiple candidate speed decision results determined by multiple total costs. Finally, for step 105, a target speed decision result is selected from the plurality of candidate speed decision results. The final target speed decision result may be selected from a plurality of candidate decision results by some screening means.
According to the method, the ST probability grid graph is established, the occupation probability in the ST probability grid graph and the relative distance between the self-vehicle and the obstacle are used in the process of calculating the total cost in the expanding process, so that the speed decision result can better avoid the grid with high occupation probability and keep away from the grid with the occupation probability as far as possible, the self-vehicle can keep a safe relative distance with the moving obstacle all the time, and the safety of the final target speed decision result can be guaranteed.
In one specific example, the ST probability grid map based on the frenet coordinate system is built as follows: establishing a current frame barrier grid map in a freset coordinate system by taking the current position of the own vehicle as an origin; determining a current frame occupied area of the obstacle in a current frame obstacle raster image according to current frame sensing data of the obstacle; and calculating and updating the current frame occupation probability of each barrier in the current frame occupation area in the current frame barrier grid image to obtain a current frame barrier risk field.
Exemplarily, the obstacle includes a dynamic obstacle, and a current frame obstacle grid map is established in a freset coordinate system with a current position of the own vehicle as an origin, which specifically includes: and establishing a current frame ST raster image corresponding to the dynamic barrier in the freset coordinate system by taking the current position of the own vehicle as an origin.
For example, determining a current occupied area of an obstacle in a current frame obstacle raster image according to current frame sensing data of the obstacle specifically includes: determining an ST area where the dynamic barrier conflicts with the self vehicle according to the current frame predicted track of the dynamic barrier and the self vehicle reference line; discretizing the ST region grid into a current frame ST grid image to obtain a first estimated occupied region; projecting the occupation area of the previous frame of the dynamic barrier to the grid image of the current frame ST to obtain a second estimated occupation area; and determining a current frame occupying area of the dynamic barrier in a current frame ST grid image according to the first estimated occupying area and the second estimated occupying area.
Illustratively, projecting the occupied area of the previous frame of the dynamic obstacle into the grid map of the current frame ST to obtain a second estimated area specifically includes: and displacing the occupied area of the previous frame of the dynamic barrier according to the time variation and the self-vehicle position variation of the previous frame and the current frame to obtain a second estimated area of the dynamic barrier in the ST raster image of the current frame.
Illustratively, calculating and updating the current frame occupation probability of each barrier in each grid in the current frame occupation area in the current frame barrier grid map to obtain a current frame barrier risk field specifically includes: calculating and updating the current frame predicted trajectory probability of each grid in the current frame occupied area of each dynamic obstacle in the current frame ST grid image; and determining the current frame occupation probability of each occupied grid in the current frame ST grid map according to the current frame predicted trajectory probability of each dynamic obstacle in the current frame occupation area of the current frame ST grid map so as to obtain a current frame ST risk field (the current frame ST risk field is the current frame ST probability grid map).
For example, determining the current frame occupation probability of each occupied grid in the current frame ST grid map according to the current frame predicted trajectory probability of each dynamic obstacle in each grid in the current frame occupation area in the current frame ST grid map specifically includes: for each occupied grid in the grid map of the current frame ST, the following steps are performed: determining at least one target dynamic barrier corresponding to each occupied grid according to the current frame occupied area of each dynamic barrier in the current frame ST grid; determining the estimated occupation probability of the current frame occupying the occupying grid according to the predicted trajectory probability of the current frame occupying the occupying grid by each target dynamic barrier and the preset maximum occupation probability; and calculating the current frame occupation probability of the occupation grid according to the estimated occupation probability of the current frame of the occupation grid and the occupation probability of the previous frame of the occupation grid.
Illustratively, the occupancy probability of each grid point in the second occupancy region is updated by using a bayesian filtering method, which is performed as follows:
illustratively, in this embodiment, a bayesian filtering algorithm may be used to update the ST risk field probability, which is as follows:
for the dynamic obstacles with perception input, updating and calculating each risk field occupying grid points P (s, t), wherein the obstacle Id is represented by i, the grid point number is represented by j, and the occupation probability of a single obstacle is represented by jiPm＝ikmNo occupation probabilityiPf＝1-ikmSince the same grid point may be occupied by multiple obstacle ST areas at the same time, the probability of the grid point being occupied is:
Pm＝min(max_p，∑ikm) (1)
whereinikmIs Id [ ]]The track probability of the dynamic obstacle with the median value of i is predicted, max _ p is the maximum occupation probability, L is an intermediate variable, and the prediction method can be obtained according to a Bayesian filter formula:
while according to the physical meaning represented by the obstacle st, when tm m-1<At 0, clear the obstacle P (S, T) risk field. The probability value of P (s, t) is high, the stability of the representing prediction result in the time dimension is high, the repeatability in the space dimension is high, a threshold interval can be defined, and the risk level of the intention of the obstacle and the corresponding prediction track are determined.
And repeating the steps to continuously update the obstacle risk field at the current moment.
And updating two parts of the computed ST risk field, namely providing accurate position information, namely, the position of the occupied grid point, and providing the occupation probability P (s, t) of each grid point, so that the establishment of an environment model of the obstacle risk field is realized, and an automatic driving system can perform decision planning based on the obstacle risk { s, t, P (s, t) }. Further, based on this occupancy probability, a speed decision plan can be made based thereon.
Further, the origin of the frenet coordinate system is a projection point of the own vehicle on the structured road, and the ordinate S is a longitudinal distance of the obstacle projected relative to the reference path of the own vehicle at the time T.
It should be noted that the specific ST probability grid map establishing process is given above only for illustrating an ST probability grid map establishing method, and is not intended to limit the scope of the present application, and any existing or future developed ST probability grid map establishing method is within the scope of the present application.
In some optional embodiments, the method further comprises: in the round-by-round expansion process, if at least two father nodes expand to the same expansion grid point, the father node with the minimum cost value is reserved. Therefore, in the process of expanding the grid points, some schemes with higher expansion cost values can be removed, some schemes with lower expansion cost values are reserved, the calculation amount is greatly reduced, and the calculation efficiency is improved.
In some alternative embodiments, the preset conditions include a speed constraint and a collision constraint. Thereby effectively avoiding overspeed and collision with other obstacles. Further optionally, the velocity constraint is a maximum velocity constraint and/or a maximum acceleration constraint; therefore, the driving safety can be better guaranteed by restricting the maximum speed and the maximum acceleration of the vehicle, for example, the maximum speed restriction and/or the maximum acceleration restriction can be set according to the speed limit regulation of the current road. In further alternative embodiments, the collision constraint is to prohibit crossing of a risk grid point, which is a grid point having an occupancy probability greater than a preset threshold. For example, the preset threshold may be set to 0.2, when the occupancy probability of a certain grid point is greater than or equal to 0.2, the certain grid point is a risk grid point, and the preset threshold may also be set to another value, which is not limited herein.
Please continue to refer to fig. 2, which shows a flowchart of another speed decision method based on probability grid map according to an embodiment of the present invention. The flow chart is further defined mainly by referring to step 103 "calculating the total cost of the path according to the occupation probability of each grid point included in the path and the relative distance between the grid point and the obstacle" in fig. 1.
As shown in fig. 2, in step 201, calculating a cost value of each expansion grid point obtained by each round of expansion according to a preset first cost function;
in step 202, the cost of the last grid point in the path is determined as the total cost of the path.
In this embodiment, after each round of expansion, the cost value of the expanded grid points may be calculated according to a preset first cost function, wherein the first cost function is configured to: the cost of the expanded grid point is the sum of the maximum value of a repulsive force function in the grid point passing from the parent node of the expanded grid point to the expanded grid point and the cost value of the parent node of the expanded grid point, and the repulsive force function is related to the occupation probability of the grid point and the relative distance between the repulsive force function and the obstacle; the cost value of the grid point at the back is obtained by superposing the cost values of the grid points at the front, so that the total cost of the path can be represented by the cost value of the last grid point, and the total cost of the current path can be directly represented by the cost value of each currently expanded grid point during expansion. That is, after each round of expansion is completed, the following steps are performed for each expansion grid point of the round of expansion: calculating the function value of the repulsion force of each grid point which passes between the expansion grid point and the father node of the expansion grid point; and taking the sum of the maximum value of the repulsion force function value and the cost value of the father node of the expansion grid point as the cost value of the expansion grid point.
With continued reference to FIG. 3, a flowchart of yet another probabilistic grid graph-based speed decision method provided by an embodiment of the invention is shown. The flow chart is mainly the flow chart further defined for step 105 "selecting a target speed decision result from the plurality of candidate speed decision results" in fig. 1.
In step 301, for each candidate speed decision result, calculating a penalty value of the candidate speed decision result according to a preset penalty function, where the penalty function is related to whether the candidate speed decision result violates a road right of an obstacle and/or destroys a historical speed decision result consistency; for example, the more times of violating the road right of the obstacle, the higher the penalty value, the more times of breaking the consistency of the historical speed decision result, the higher the penalty value.
In step 302, a target speed decision result is selected from the plurality of candidate speed decision results according to the penalty values of the plurality of candidate speed decision results and the size of s projected by the paths corresponding to the penalty values in the ST probability grid map.
In this embodiment, after the velocity decision device based on the probability grid map calculates the cost of expanding grid points according to the repulsive potential function and then obtains a plurality of candidate velocity decision results according to the continuous expansion result, a target velocity decision result needs to be selected from a plurality of feasible decision results. Therefore, the present embodiment mainly calculates a penalty value of the candidate speed decision result according to a preset penalty function, wherein the penalty function is related to whether the candidate speed decision result violates the road weight of the obstacle and/or destroys the historical speed decision result consistency. And finally, comprehensively considering the penalty value and the size of a vertical coordinate s corresponding to a plurality of paths of the feasible speed decision result in the ST probability grid diagram, and determining a final speed decision result.
The road right mainly refers to an obstacle with a higher road right as much as possible according to the traffic rule requirement, the consistency determination mainly determines whether the current speed decision result is consistent with the historical speed decision result in a decision pool (a previously executed speed decision result or a decision pool formed by previous behaviors), for example, the current speed decision result is at least consistent with the latest speed decision result for the same obstacle, so that the overall vehicle performance is not hesitant and unstable, or the consistency determination for the same obstacle is set to be consistent with the latest N decision results, for example, the latest 3 decision results for the obstacle a in the decision pool are all yield, if the current decision result is yield, the current decision result is consistent with the previous decision result, if the current decision result is overtaking, the current decision result is inconsistent with the previous decision result, the penalty value calculation needs to be performed through a penalty function, the application is not limited herein.
In a specific example, for consistency, for example, if the historical speed decision result yields for 2 obstacles, if the current decision result yields one obstacle more than one obstacle, the decision result may be recorded as consistency violation once, the penalty value of consistency is recorded as 1, and if the current decision result exceeds 2 obstacles, the decision result may be recorded as consistency violation twice, and each obstacle decision result change may be recorded as consistency violation once, which is applicable to a plurality of obstacles, and details are not described herein. For the road right, a road right is violated once, and the penalty value is recorded as 1, which is not described herein again. The road weight and the consistency can also have corresponding coefficients, so that the calculation result is more reliable, and the determination of the coefficients is not described herein again.
As a further alternative, please refer to fig. 4, which shows a flowchart of a further limited step of "selecting a target speed decision result from the plurality of candidate speed decision results according to the penalty values of the plurality of candidate speed decision results and the size of s projected by the paths corresponding to the penalty values in the ST probability grid map" in step 302.
As shown in fig. 4, in step 401, a candidate speed decision result with the lowest penalty value is selected;
in step 402, if there is one candidate speed decision result with the lowest penalty value, determining the candidate speed decision result with the lowest penalty value as a target speed decision result;
in step 403, if there are a plurality of speed candidate decision results with the lowest penalty value, the speed candidate decision result corresponding to the path with the largest ordinate s in the ST probability grid map is selected and determined as the target speed decision result.
The method of the embodiment mainly characterizes how to obtain a final target speed decision result according to the penalty value, when only one minimum penalty value exists, the candidate speed decision result corresponding to the minimum penalty value can be directly determined as the final target decision result, when a plurality of minimum penalty values exist, the size of a vertical coordinate s in an ST probability grid diagram needs to be further considered, and the larger the vertical coordinate s is, the higher the driving efficiency is, and the safer the driving efficiency is.
Referring to fig. 5, a flowchart of a speed decision method based on a probability grid map according to another embodiment of the present invention is shown. The flowchart is mainly a flowchart further defined by step 103 "calculating the total cost of the path according to the occupation probability of each grid point included in the path and the relative distance between the grid point and the obstacle" in fig. 1.
As shown in fig. 5, in step 501, the cost of each expansion grid point obtained by each round of expansion is calculated according to a preset second cost function; wherein the second cost function is configured to obtain a repulsive force potential function, a penalty function for violating the road weight of the obstacle and/or destroying consistency, and a cost function obtained by comprehensively weighting the acceleration of the vehicle; the repulsive force potential function is related to the probability of occupation of a grid point and its relative distance to an obstacle. That is, after each round of expansion is completed, for each expanded grid point of the round of expansion, the following steps are performed: calculating a repulsive force potential function value of each grid point passing between the expansion grid point and the father node of the expansion grid point; calculating the road right of the path where the expansion grid point is positioned violating the barrier and/or the penalty value of breaking consistency according to a preset penalty function; and taking the maximum value of the repulsive force potential function value, the weighted sum of the punishment value and the acceleration of the self vehicle as the cost value of the expanded grid point.
In step 502, the cost of the last grid point on the path is determined as the total cost of the path.
The second cost function of the present embodiment is different from the first cost function in that the second cost function is a cost function obtained by comprehensively weighting the repulsive force potential function, the penalty function violating the road weight of the obstacle and/or breaking the consistency, and the self-vehicle acceleration, so that the two functions are integrated into the second cost function without separate repulsive force potential function and penalty function, and a final speed decision result can be obtained by using one cost function, and the calculation efficiency is higher.
It should be noted that for simplicity of explanation, the foregoing method embodiments are described as a series of acts or combination of acts, but those skilled in the art will appreciate that the present invention is not limited by the order of acts, as some steps may occur in other orders or concurrently in accordance with the invention. Further, those skilled in the art should also appreciate that the embodiments described in the specification are preferred embodiments and that the acts and modules referred to are not necessarily required by the invention. In the foregoing embodiments, the descriptions of the respective embodiments have respective emphasis, and for parts that are not described in detail in a certain embodiment, reference may be made to related descriptions of other embodiments.
Further, a calculation formula of the repulsive force potential function mentioned in the above embodiment is as follows:
wherein, Ureq(Q) represents repulsive force of obstacle around grid point Q on grid point Q in ST probability grid diagram, eta is repulsive gain, Q*Is a range threshold of the obstacle, the range threshold Q*In relation to the vehicle speed of the own vehicle, p is an occupation probability of a grid point q, and d (q) is a relative distance of the grid point q with respect to the own vehicle. The repulsion potential function is improved by combining the probability in the ST probability grid graph, when the relative distance is larger than the action distance threshold value, the safe driving distance between the vehicle and the obstacle is kept, the repulsion potential is 0 at the moment, when the relative distance is smaller than or equal to the action distance threshold value, the repulsion potential is generated, the repulsion potential is in negative correlation with the relative distance, namely the repulsion potential is larger when the relative distance is smaller, the repulsion potential is in positive correlation with the occupation probability, namely the repulsion potential is larger when the occupation probability is higher. The repulsive force between the moving barrier and the characteristic can be better represented, so that the expanded path can avoid the grid with high occupation probability and be far away from the grid with occupation probability as far as possible, and a safer path is selected for final speed decision.
It should be noted that, when calculating the repulsive force potential function of each grid point, because the grid point may be occupied by different obstacles, that is, the calculation results of the repulsive force potential functions of different obstacles for the grid point are different, the maximum value of each repulsive force potential function at the grid point may be taken when calculating the repulsive force potential function, so that the grid point is the maximum value of the repulsive force potential functions of each obstacle as the epoch value of the expanded grid point, thereby better characterizing the danger of the grid point and allowing the self-vehicle to better avoid the grid point with the danger. In other schemes, a plurality of obstacles (all satisfying relative distance greater than Q) can be further processed*) And performing weighted calculation on each repulsive force function of a certain grid point to obtain the value of the repulsive force function of the grid point, which is not limited in the application.
For a better understanding of the application's improvement to the repulsive potential function by those skilled in the art, the original repulsive potential function is now given for reference: the basic idea of the artificial potential field method is that a simulated barrier generates repulsion force on the motion of a self-vehicle, the self-vehicle avoids the barrier under the guidance of a potential field, and the traditional repulsion force potential function is as follows:
wherein D (Q) is the distance between the point Q and the nearest obstacle, eta is the repulsive force gain, Q*Is the threshold of the acting distance of the obstacle, and the obstacle larger than the acting distance does not generate the influence of repulsion.
The inventor finds in the process of implementing the present application that since the probabilistic grid map does not provide the exact location of an obstacle, but provides the probability of occupancy, the repulsive force potential function of a grid point can be calculated from the relative distances and occupancy probabilities of other grid points.
Please refer to fig. 6, which illustrates a schematic diagram of the speed decision result in an ST probability grid chart according to an embodiment of the present invention. In the following, the calculation of penalty values based on consistency and road weights and the process of determining the final speed decision result are given according to the diagram. Where the abscissa in fig. 6 is time T, the ordinate is S, and 0.2, 0.4, and 0.6 labeled within a grid point in the figure are probabilities that an obstacle occupies the grid point in the ST probability grid map.
There may be multiple topologies in the multiple feasible decision results generated by the aforementioned expanding step search, that is, there are multiple permutation and combination of the decision results for each obstacle. For example, in fig. 6, there are three total topology choices, namely, first to overtake two obstacles, second to overtake the first and yield the second obstacle, and fourth to yield the two obstacles. The same decision result is usually made on the same barrier in the driving process of the vehicle, otherwise, the overall performance of the vehicle is hesitant and unstable, and meanwhile, the barrier with higher road right is given as much as possible according to the requirement of traffic rules. Therefore, a speed decision result which is as consistent as possible with the historical decision result and accords with the road right needs to be selected from all feasible solutions, and the penalty term of the decision result is designed to be
E＝αm+βn，
Wherein m and n are the times of destroying consistency and road weight respectively, and alpha and beta are coefficients of two punishment terms respectively. For example, the right of way of the first obstacle is higher than that of the own vehicle, and the right of way of the second obstacle is lower than that of the own vehicle, and the decision result at the last moment is to overtake the first obstacle and give way to the second obstacle. Under the conditions that alpha is 2 and beta is 1, the penalty value of the feasible solution is 3, the penalty values of the feasible solution are 1, the penalty value of the feasible solution is 2, and therefore the topology corresponding to the feasible solution is selected. Considering that the larger the final search is, the higher the driving efficiency is, the feasible solution is taken as the final speed decision result.
As a variation, the cost function for calculating the total cost according to the embodiment of the present invention may be configured to be related to more parameters, for example, may be related to one or more of the probability that each grid point is occupied by an obstacle at each time, the relative distance between the vehicle and the obstacle, the vehicle speed acceleration of the vehicle, whether the current speed decision result of the vehicle is consistent with the historical decision result in the decision pool, and the broken road right, and it is understood that the more parameters, the faster the final speed decision result can be obtained.
In some optional embodiments, the cost function is a cost function obtained by comprehensively weighting a repulsive force potential function, a penalty function violating the road weight and/or consistency of the obstacle, and the acceleration, wherein the cost function is in positive correlation with the repulsive force potential function, the penalty function, and the acceleration. Therefore, various parameters can be weighted to form a cost function, and the final decision result is obtained by utilizing the cost function.
With continued reference to FIG. 7, a flowchart of an overall feasible solution search based on probability grid graphs provided by an embodiment of the invention is shown.
Wherein the ST probability grid map reflects the probability of each position on the own vehicle planning path being occupied at each time point, wherein grid points with the occupation probability greater than a certain threshold are regarded as grid points which are forbidden to pass through. The speed decision needs to avoid all grids forbidden to pass through and simultaneously keep away from grids with occupation probability as far as possible, so that the potential function is used as a cost function to search a feasible solution of the speed decision.
The flow of the particular speed decision feasible solution search is shown in fig. 7. The search step is a fixed time, for example, 1 second, the initial grid point, i.e., the parent node, is (0, 0), the expanded grid point needs to satisfy the constraints of speed limit and maximum acceleration, and a risk grid point which is forbidden to pass through cannot pass through with the parent node, and the cost value of the expanded grid point is the maximum value of a repulsive potential function in the passed grid point. And obtaining grid points which are possibly reached in 1s after the first-step search is completed, then sequentially using the grid points as father nodes to continue expansion, similarly, the requirements of kinematic constraint and collision constraint are met, at the moment, the total cost value of the expanded grid points is the sum of the cost value of the father nodes and the maximum value of the potential function in the passed grid points, and if other father nodes expand to the same grid point and the total cost value is smaller, the father nodes and the total cost value of the expanded grid points are updated. And stopping searching when the searching length reaches the planning time length, and obtaining a plurality of feasible searching results by reversely searching the father node.
The inventor discovers that in the process of implementing the application: because speed decision and planning are very dependent on the results of environment perception and prediction, real and stable distribution and intention of other traffic participants need to be obtained, and the probability grid map can effectively combine time and space information, the measured positions of the obstacles are converted into the probability of being occupied by each position, and the influence of input disturbance is relieved to a certain extent. Therefore, the inventor provides a speed decision method based on the probability grid map, a plurality of feasible decision results are searched in the probability grid map according to the repulsive force field function, the influence of uncertainty of perception and prediction information can be reduced, meanwhile, the historical decision results and the relative road weight relationship are referred, and the stability and the reasonability of the decision results are guaranteed.
Referring to fig. 8, a block diagram of a velocity decision device based on a probability grid map according to an embodiment of the invention is shown.
As shown in FIG. 8, the probabilistic grid graph-based speed decision device 800 includes a probabilistic grid graph creation module 810, an expansion module 820, a total cost calculation module 830, a candidate determination module 840, and a decision module 850.
The probability grid map establishing module 810 is configured to establish an ST probability grid map in a friends coordinate system with a self-parking location as an origin; an expansion module 820, configured to expand, according to a search step length, a plurality of expansion grid points that satisfy a preset condition, in the ST probability grid map, with a grid point where the own vehicle is located as a parent node; continuously expanding a plurality of expanded grid points which meet preset conditions according to the search step length by taking each expanded grid point as a father node; searching in turn until a preset termination condition is met to obtain a plurality of paths and corresponding speed decision results; a total cost calculation module 830, configured to calculate, for each path, a total cost of the path according to an occupation probability of each grid point included in the path and a relative distance between the path and an obstacle; a candidate determining module 840, configured to determine a plurality of candidate speed decision results according to the total costs corresponding to the plurality of paths; and a decision module 850 for selecting a target speed decision result from the plurality of candidate speed decision results.
In some optional embodiments, the above speed decision device 800 based on a probability grid map further includes a retaining module (not shown in the figure) for retaining the parent node with the smallest cost value if there are at least two parent nodes extending to the same extended grid point in the round-by-round extension process.
Optionally, the preset condition includes a speed constraint and a collision constraint.
Further optionally, the velocity constraint is a maximum velocity constraint and/or a maximum acceleration constraint; the collision constraint is to forbid crossing of a risk grid point, wherein the risk grid point is a grid point with an occupation probability greater than a preset threshold.
In some optional embodiments, the total cost calculating module 830 specifically includes instructions for: calculating the cost value of each expansion grid point obtained by each round of expansion according to a preset first cost function; wherein the first cost function is configured to: the cost of the expanded grid point is the sum of the maximum value of a repulsive force function in the grid point passing from the parent node of the expanded grid point to the expanded grid point and the cost value of the parent node of the expanded grid point, and the repulsive force function is related to the occupation probability of the grid point and the relative distance between the repulsive force function and the obstacle; and determining the cost of the last grid point in the path as the total cost of the path.
In some alternative embodiments, the decision module 850 specifically includes instructions for: calculating a penalty value of each candidate speed decision result according to a preset penalty function, wherein the penalty function is related to whether the candidate speed decision result violates the road right of the obstacle and/or destroys the consistency of historical speed decision results; and selecting a target speed decision result from the candidate speed decision results according to the punishment values of the candidate speed decision results and the projection s of the paths corresponding to the punishment values in the ST probability grid map.
Further optionally, selecting a target speed decision result from the multiple candidate speed decision results according to penalty values of the multiple candidate speed decision results and a size of a projection s of the penalty values in the ST probability grid map, where the selecting specifically includes: selecting a candidate speed decision result with the lowest penalty value; if the candidate speed decision result with the lowest penalty value is one, determining the candidate speed decision result with the lowest penalty value as a target speed decision result; and if a plurality of candidate speed decision results with the lowest penalty values are obtained, selecting the candidate speed decision result corresponding to the path with the maximum vertical coordinate s in the ST probability grid map to be determined as the target speed decision result.
In other alternative embodiments, the total cost calculating module 830 specifically includes instructions for: calculating the cost of each expansion grid point obtained by each round of expansion according to a preset second cost function; wherein the second cost function is configured to obtain a repulsive force potential function, a penalty function of violating road weight of the obstacle and/or breaking consistency, and a cost function obtained by comprehensively weighting the acceleration of the vehicle; the repulsive force potential function is related to the occupation probability of the grid points and the relative distance between the grid points and the obstacle; and determining the cost of the last grid point on the path as the total cost of the path.
In some embodiments, the present invention provides a non-transitory computer readable storage medium, in which one or more programs including executable instructions are stored, and the executable instructions can be read and executed by an electronic device (including but not limited to a computer, a server, or a network device, etc.) to perform any one of the above-mentioned probabilistic grid graph-based speed decision methods of the present invention.
In some embodiments, embodiments of the invention also provide a computer program product comprising a computer program stored on a non-transitory computer readable storage medium, the computer program comprising program instructions that, when executed by a computer, cause the computer to perform any of the above probability grid map based speed decision methods.
In some embodiments, an embodiment of the present invention further provides an electronic device, which includes: at least one processor, and a memory communicatively coupled to the at least one processor, wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform a probabilistic raster graph-based speed decision method.
In some embodiments, the present invention further provides a mobile device, including a mobile body and the electronic apparatus described in the foregoing embodiments. Wherein the mobile device may be: vehicles, robots, etc. The vehicle may be an unmanned vehicle, such as an unmanned sweeper, an unmanned ground washer, an unmanned logistics vehicle, and an unmanned taxi.
Fig. 9 is a schematic hardware structure diagram of an electronic device for performing a velocity decision method based on a probability grid map according to another embodiment of the present application, and as shown in fig. 9, the device includes:
one or more processors 910 and memory 920, one processor 910 being exemplified in fig. 9.
The apparatus for performing the probabilistic grid graph-based speed decision method may further include: an input device 930 and an output device 940.
The processor 910, the memory 920, the input device 930, and the output device 940 may be connected by a bus or other means, and fig. 9 illustrates an example of a connection by a bus.
The memory 920, which is a non-volatile computer-readable storage medium, may be used to store non-volatile software programs, non-volatile computer-executable programs, and modules, such as program instructions/modules corresponding to the speed decision method based on probability grid maps in the embodiments of the present application. The processor 910 executes various functional applications of the server and data processing by running nonvolatile software programs, instructions and modules stored in the memory 920, i.e. implementing the speed decision method based on the probability grid map of the above method embodiment.
The memory 920 may include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function; the storage data area may store data created from use of a probability grid map-based speed decision device, and the like. Further, the memory 920 may include high speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid state storage device. In some embodiments, the memory 920 may optionally include memory located remotely from the processor 910, which may be connected to the probability grid map based speed decision device via a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
The input device 930 may receive input numeric or character information and generate signals related to user settings and function control of the probability grid map-based speed decision device. The output device 940 may include a display device such as a display screen.
The one or more modules are stored in the memory 920 and, when executed by the one or more processors 910, perform a probabilistic grid graph-based speed decision method in any of the method embodiments described above.
The product can execute the method provided by the embodiment of the application, and has the corresponding functional modules and beneficial effects of the execution method. For technical details that are not described in detail in this embodiment, reference may be made to the methods provided in the embodiments of the present application.
The electronic device of the embodiments of the present application exists in various forms, including but not limited to:
(1) mobile communication devices, which are characterized by mobile communication capabilities and are primarily targeted at providing voice and data communications. Such terminals include smart phones, multimedia phones, functional phones, and low-end phones, among others.
(2) The ultra-mobile personal computer equipment belongs to the category of personal computers, has calculation and processing functions and generally has the characteristic of mobile internet access. Such terminals include PDA, MID, and UMPC devices, among others.
(3) Portable entertainment devices such devices may display and play multimedia content. The devices comprise audio and video players, handheld game consoles, electronic books, intelligent toys and portable vehicle-mounted navigation devices.
(4) Other onboard electronic devices with data interaction functions, such as a vehicle-mounted device mounted on a vehicle.
The above-described embodiments of the apparatus are merely illustrative, and the units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one position, or may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment.
Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented by software plus a general hardware platform, and certainly can also be implemented by hardware. Based on such understanding, the above technical solutions substantially or contributing to the related art may be embodied in the form of a software product, which may be stored in a computer-readable storage medium, such as ROM/RAM, magnetic disk, optical disk, etc., and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method according to the embodiments or some parts of the embodiments.
Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit the same; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions in the embodiments of the present application.
Claims (14)
1. A probabilistic grid graph-based speed decision method, the method comprising:
establishing an ST probability grid map in a freset coordinate system by taking a self-parking position as an origin;
expanding a plurality of expanded grid points meeting preset conditions according to search step length by taking the grid point where the self-parking position is as a father node in the ST probability grid graph; taking each expansion grid point as a father node, and continuing to expand a plurality of expansion grid points meeting preset conditions according to the search step length; searching round by round until a preset termination condition is met to obtain a plurality of paths and corresponding speed decision results;
for each path, calculating the total cost of the path according to the occupation probability of each grid point contained in the path and the relative distance between the path and an obstacle;
determining a plurality of candidate speed decision results according to the total cost corresponding to the plurality of paths;
and selecting a target speed decision result from the candidate speed decision results.
2. The method of claim 1, further comprising:
in the round-by-round expansion process, if at least two father nodes expand to the same expansion grid point, the father node with the minimum cost value is reserved.
3. The method of claim 1, wherein the preset conditions include a speed constraint and a collision constraint.
4. A method according to claim 3, wherein the velocity constraint is a maximum velocity constraint and/or a maximum acceleration constraint;
the collision constraint is to forbid crossing of a risk grid point, wherein the risk grid point is a grid point with an occupation probability greater than a preset threshold.
5. The method according to claim 1, wherein calculating the total cost of the path according to the occupation probability of each grid point included in the path and the relative distance between the grid point and the obstacle comprises:
calculating the cost value of each expansion grid point obtained by each round of expansion according to a preset first cost function; wherein the first cost function is configured to: the cost of the expanded grid point is the sum of the maximum value of a repulsive force function in the grid point passing from the parent node of the expanded grid point to the expanded grid point and the cost value of the parent node of the expanded grid point, and the repulsive force function is related to the occupation probability of the grid point and the relative distance between the repulsive force function and the obstacle;
determining a cost of a last grid point in the path as a total cost of the path.
6. The method according to any one of claims 1-5, wherein selecting a target speed decision result from the plurality of candidate speed decision results specifically comprises:
calculating a penalty value of each candidate speed decision result according to a preset penalty function, wherein the penalty function is related to whether the candidate speed decision result violates the road right of an obstacle or not and/or the consistency of the damage historical speed decision result;
and selecting a target speed decision result from the candidate speed decision results according to the punishment values of the candidate speed decision results and the projection s of the paths corresponding to the punishment values in the ST probability grid map.
7. The method as claimed in claim 6, wherein selecting a target speed decision result from the plurality of candidate speed decision results according to penalty values of the plurality of candidate speed decision results and a size of a projection s thereof in the ST probability grid map comprises:
selecting a candidate speed decision result with the lowest penalty value;
if the candidate speed decision result with the lowest penalty value is one, determining the candidate speed decision result with the lowest penalty value as a target speed decision result;
and if a plurality of candidate speed decision results with the lowest penalty values are obtained, selecting the candidate speed decision result corresponding to the path with the maximum vertical coordinate s in the ST probability grid map to be determined as the target speed decision result.
8. The method according to claim 1, wherein calculating the total cost of the path according to the occupation probability of each grid point included in the path and the relative distance between the grid point and the obstacle comprises:
calculating the cost of each expansion grid point obtained by each round of expansion according to a preset second cost function; the second cost function is configured to be a cost function obtained by comprehensively weighting a repulsive force potential function, a road weight violating an obstacle and/or a penalty function destroying the consistency of historical speed decision results, and the acceleration of the vehicle; the repulsive force potential function is related to the occupation probability of the grid points and the relative distance between the grid points and the obstacle;
determining the cost of the last grid point on the path as the total cost of the path.
9. The method according to claim 5 or 8, wherein the repulsive potential function is calculated as follows:
wherein, Ureq(Q) represents repulsive force of obstacle around grid point Q on grid point Q in ST probability grid diagram, eta is repulsive gain, Q*Is a range threshold of the obstacle, the range threshold Q*In relation to the vehicle speed, p is an occupancy probability of a grid point q, and d (q) is a relative distance between the vehicle and the grid point q.
10. A probabilistic grid-graph-based speed decision device, comprising:
the probability grid map establishing module is used for establishing an ST probability grid map in a freset coordinate system by taking a self-parking position as an origin;
the expansion module is used for expanding a plurality of expansion grid points meeting preset conditions according to search step length by taking the grid point where the self-parking position is located as a father node in the ST probability grid graph; taking each expansion grid point as a father node, and continuing to expand a plurality of expansion grid points meeting preset conditions according to the search step length; searching round by round until a preset termination condition is met to obtain a plurality of paths and corresponding speed decision results;
the total cost calculation module is used for calculating the total cost of each path according to the occupation probability of each grid point contained in the path and the relative distance between the path and an obstacle;
the candidate determining module is used for determining a plurality of candidate speed decision results according to the total cost corresponding to the paths;
and the decision module is used for selecting a target speed decision result from the candidate speed decision results.
11. An electronic device, comprising: at least one processor, and a memory communicatively coupled to the at least one processor, wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the steps of the method of any one of claims 1-9.
12. A mobile device comprising a body and the electronic apparatus of claim 11 mounted on the body.
13. A storage medium on which a computer program is stored which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 9.
14. A computer program product comprising a computer program, characterized in that the computer program realizes the steps of the method of any one of claims 1-9 when executed by a processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210272793.7A CN114633765A (en) | 2022-03-18 | 2022-03-18 | Speed decision method and device based on probability grid graph and related products |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210272793.7A CN114633765A (en) | 2022-03-18 | 2022-03-18 | Speed decision method and device based on probability grid graph and related products |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114633765A true CN114633765A (en) | 2022-06-17 |
Family
ID=81950213
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210272793.7A Pending CN114633765A (en) | 2022-03-18 | 2022-03-18 | Speed decision method and device based on probability grid graph and related products |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114633765A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115782867A (en) * | 2022-11-17 | 2023-03-14 | 上海西井信息科技有限公司 | Track collision risk assessment method and device, electronic equipment and storage medium |
-
2022
- 2022-03-18 CN CN202210272793.7A patent/CN114633765A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115782867A (en) * | 2022-11-17 | 2023-03-14 | 上海西井信息科技有限公司 | Track collision risk assessment method and device, electronic equipment and storage medium |
| CN115782867B (en) * | 2022-11-17 | 2024-01-30 | 上海西井科技股份有限公司 | Track collision risk assessment method, device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11900797B2 (en) | Autonomous vehicle planning | |
| WO2022052406A1 (en) | Automatic driving training method, apparatus and device, and medium | |
| KR102325028B1 (en) | Method and device for performing multiple agent sensor fusion in cooperative driving based on reinforcement learning | |
| CN111898211B (en) | Intelligent vehicle speed decision method based on deep reinforcement learning and simulation method thereof | |
| US11648935B2 (en) | Driving assistance method and system | |
| Lin et al. | Decision making through occluded intersections for autonomous driving | |
| CN112249032B (en) | Automatic driving decision method, system, equipment and computer storage medium | |
| KR20200096115A (en) | Method and device for short-term path planning of autonomous driving through information fusion by using v2x communication and image processing | |
| CN109213153B (en) | Automatic vehicle driving method and electronic equipment | |
| Schörner et al. | Predictive trajectory planning in situations with hidden road users using partially observable markov decision processes | |
| US20220097736A1 (en) | Vehicle Control Method and Apparatus, Storage Medium, and Electronic Device | |
| González et al. | High-speed highway scene prediction based on driver models learned from demonstrations | |
| CN109115220B (en) | Method for parking lot system path planning | |
| CN114894206A (en) | Path planning method and device, vehicle and storage medium | |
| CN114644016A (en) | Vehicle automatic driving decision-making method and device, vehicle-mounted terminal and storage medium | |
| CN114281084A (en) | Intelligent vehicle global path planning method based on improved A-x algorithm | |
| CN114633765A (en) | Speed decision method and device based on probability grid graph and related products | |
| CN113139696B (en) | Trajectory prediction model construction method and trajectory prediction method and device | |
| CN114620071A (en) | Detour trajectory planning method, device, equipment and storage medium | |
| CN114782912A (en) | Obstacle risk field environment modeling method and device and related products | |
| CN114723903A (en) | Obstacle risk field environment modeling method and device and related products | |
| KR20230024392A (en) | Driving decision making method and device and chip | |
| Schörner et al. | Towards Multi-Modal Risk Assessment | |
| CN116142230A (en) | Speed planning method and device, control equipment, vehicle and storage medium | |
| US20240092385A1 (en) | Driving Policy Determining Method and Apparatus, Device, and Vehicle |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information |
Address after: Room 101-901, 8th Floor, Building 4, Zone 3, No. 22, Jinghai 4th Road, Beijing Economic and Technological Development Zone, Daxing District, Beijing 100176 (Yizhuang Cluster, High end Industrial Zone, Beijing Pilot Free Trade Zone) Applicant after: Beijing Idriverplus Technology Co.,Ltd. Address before: 100176 room 2602, 22 / F, building 4, yard 8, Wenhua Park West Road, Beijing Economic and Technological Development Zone, Daxing District, Beijing Applicant before: Beijing Idriverplus Technology Co.,Ltd. |
|
| CB02 | Change of applicant information |