CN111746523A - Vehicle parking path planning method and device, vehicle and storage medium - Google Patents

Vehicle parking path planning method and device, vehicle and storage medium Download PDF

Info

Publication number
CN111746523A
CN111746523A CN202010608892.9A CN202010608892A CN111746523A CN 111746523 A CN111746523 A CN 111746523A CN 202010608892 A CN202010608892 A CN 202010608892A CN 111746523 A CN111746523 A CN 111746523A
Authority
CN
China
Prior art keywords
vehicle
parking path
path planning
obstacle
collision avoidance
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
Application number
CN202010608892.9A
Other languages
Chinese (zh)
Inventor
刘鑫
王维
贺志国
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Sany Special Vehicle Co Ltd
Original Assignee
Sany Special Vehicle Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Sany Special Vehicle Co Ltd filed Critical Sany Special Vehicle Co Ltd
Priority to CN202010608892.9A priority Critical patent/CN111746523A/en
Publication of CN111746523A publication Critical patent/CN111746523A/en
Pending legal-status Critical Current

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
    • B60W30/00—Purposes of road vehicle drive control systems not related to the control of a particular sub-unit, e.g. of systems using conjoint control of vehicle sub-units
    • B60W30/06—Automatic manoeuvring for parking
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G06Q10/047—Optimisation of routes or paths, e.g. travelling salesman problem

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • Strategic Management (AREA)
  • Economics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Operations Research (AREA)
  • Mechanical Engineering (AREA)
  • Game Theory and Decision Science (AREA)
  • Transportation (AREA)
  • Automation & Control Theory (AREA)
  • Marketing (AREA)
  • Development Economics (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Control Of Driving Devices And Active Controlling Of Vehicle (AREA)

Abstract

The invention provides a vehicle parking path planning method and device, a vehicle and a storage medium. The vehicle parking path planning method comprises the following steps: constructing a kinematic model of the vehicle according to the size parameters and the motion state of the vehicle; according to the distribution condition of obstacles around the vehicle, constructing an obstacle model of the vehicle, and acquiring a collision avoidance constraint condition of the vehicle and the obstacles; reconstructing the collision avoidance constraints so that the collision avoidance constraints are continuous; representing the kinematic model and constructing a target model function; selecting an initial value of the target model function to obtain an initial value selection result; and selecting a result according to the initial value, and solving the target model function to obtain a parking path planning result of the vehicle. The invention can plan a smooth and natural vehicle parking path without collision and satisfying motion constraint.

Description

Vehicle parking path planning method and device, vehicle and storage medium
Technical Field
The invention relates to the technical field of automatic parking, in particular to a vehicle parking path planning method, a vehicle parking path planning device, a vehicle and a storage medium.
Background
The path planning is an important technology in vehicle motion planning, and in the current path planning research process of autonomous parking, a path planning algorithm usually adopts a mode of searching first and then optimizing, and aims to make a trajectory curve smooth and have continuous second order. The related art lacks a technical solution capable of obtaining a vehicle parking path that is smooth and satisfies motion constraints.
Disclosure of Invention
The present invention is directed to solving at least one of the above problems.
To this end, a first object of the present invention is to provide a vehicle parking path planning method.
A second object of the present invention is to provide a vehicle.
A third object of the present invention is to provide a vehicle parking path planning apparatus.
A fourth object of the present invention is to provide a computer-readable storage medium.
To achieve the first object of the present invention, an embodiment of the present invention provides a vehicle parking path planning method, including: constructing a kinematic model of the vehicle according to the size parameters and the motion state of the vehicle; according to the distribution condition of obstacles around the vehicle, constructing an obstacle model of the vehicle, and acquiring a collision avoidance constraint condition of the vehicle and the obstacles; reconstructing the collision avoidance constraints so that the collision avoidance constraints are continuous; representing the kinematic model and constructing a target model function; selecting an initial value of the target model function to obtain an initial value selection result; and selecting a result according to the initial value, and solving the target model function to obtain a parking path planning result of the vehicle.
The vehicle parking path planning method of the embodiment can generate a smooth path which meets the vehicle kinematics model and has no collision. In addition, the trajectory generated by the vehicle parking path planning method of the present embodiment is smoother and easier to track.
In addition, the technical solution provided by the above embodiment of the present invention may further have the following additional technical features:
in the above technical solution, in constructing a kinematic model of the vehicle according to the dimensional parameters and the motion state of the vehicle, the kinematic model includes:
Figure RE-RE-GDA0002614535890000021
wherein a two-dimensional plane rectangular coordinate system is constructed based on the central axis of the vehicle as the X axis, (X, Y) is the central position coordinate between two rear wheels of the vehicle,
Figure RE-RE-GDA0002614535890000022
is the angle between the front wheels of the vehicle and the X-axis, v is the speed of the vehicle in the direction of the central axis, is the steering angle of the vehicle, L is the distance between the front and rear wheels of the vehicle, and α is the acceleration of the vehicle.
The present embodiment is modeled by a vehicle dynamics single-vehicle model, and the state of the vehicle 100 is represented by the above model formula in a low speed state.
In any of the above technical solutions, in the process of constructing an obstacle model of a vehicle according to the distribution of obstacles around the vehicle and obtaining the collision avoidance constraint conditions between the vehicle and the obstacles,
the obstacle model includes:
O(m)＝{A(m)y≤b(m)|y∈R2}；
wherein A is(m)And b(m)Respectively representing known conditions, O, relating to the distribution of said obstacles(m)Representing the multicellular shape of the obstacle, R representing the real number domain, R2A real number domain representing a two-dimensional space, y representing a y coordinate point in a coordinate point function (x, y) in the real number domain;
the collision avoidance constraints include:
Figure RE-RE-GDA0002614535890000023
wherein the content of the first and second substances,e (x) denotes the multicellular shape of the vehicle, O(m)Represents the shape of multiple cells of an obstacle in which E (x) and O are in the collision avoidance constraint(m)The intersection of (a) and (b) indicates that the vehicle does not collide with the obstacle, M indicates the number of obstacles, and M indicates the current mth obstacle.
The obstacle model of the present embodiment is applicable to most obstacles, which may be approximated as a union of polyhedrons.
In any of the above technical solutions, in reconstructing the collision avoidance constraint condition so that the collision avoidance constraint condition has continuity, the reconstructed collision avoidance constraint condition includes:
sd(E(x),O(m))：＝dist(E(x),O(m))-pen(E(x),O(m))；
where dist is a function of the distance between the vehicle and the obstacle, pen is a function of the penetration between the vehicle and the obstacle, sd (E (x), O(m)) Represents the calculation of E (x) ∩ O(m)To determine whether the vehicle collides with an obstacle, sd (E (x), O(m)) If the result of (2) is greater than 0, E (x) ∩ O is determined(m)Is equal to
Figure RE-RE-GDA0002614535890000031
Obtaining sd (E (x), O(m)) If the result of (3) is less than or equal to 0, E (x) ∩ O is determined(m)Is not equal to
Figure RE-RE-GDA0002614535890000032
In the above manner, the collision avoidance constraint is reconstructed based on the concept of the symbolic distance function, so as to ensure the continuity of the collision avoidance constraint.
In any of the above technical solutions, in representing a kinematic model and constructing an objective model function,
the representation of the kinematic model includes:
xk+1＝xk+τf(xk+0.5τf(xk,uk),uk)；
wherein, tau is the sampling timeAnd τ is greater than 0, xk+1Is the state variable, x, of the vehicle at time k +1kIs the state variable of the vehicle at time k, ukThe control variable of the vehicle at the moment k; f (x)k+ 0.5τf(xk,uk),uk) Representing a motion state of the vehicle, the motion state including a center position coordinate (x, y) between two rear wheels of the vehicle, an angle between a front wheel of the vehicle and an x-axis
Figure RE-RE-GDA0002614535890000033
And a speed v of the vehicle in the center axis direction;
the objective model function includes:
Figure RE-RE-GDA0002614535890000034
wherein x issRepresenting the starting state of said vehicle, xFRepresenting the final state of the vehicle, R (x)k) And t (x)k) Respectively representing a rotation matrix and a translation matrix of said vehicle, G and G representing known matrices associated with a starting state of said vehicle, λ and μ being auxiliary decision variables,
Figure RE-RE-GDA0002614535890000035
and
Figure RE-RE-GDA0002614535890000036
is an obstacle O at time k(m)Of a decision variable, τ>0 is the sampling time, xkAnd ukFor the state variables and control variables, Q and Q, of the vehicle at time kΔFor a semi-positive weighting matrix, Deltau, of said vehicle with respect to a control quantityk＝(uk-uk-1) K is the weight that trades off the minimum time, h (x)k，uk) The state constraint of the vehicle is less than or equal to 0,
Figure RE-RE-GDA0002614535890000041
μ≥0：-gTμ+(At(x)-b)Tλ>d，GTμ+R(x)TATλ＝0， ||ATλ | | ═ 1 is the reconstructed collision avoidance constraint.
In this embodiment, a kinematic model of a vehicle is represented, and the target model function is further constructed, so that an open source solver can solve the target model function.
In any of the above technical solutions, selecting an initial value of the target model function to obtain an initial value selection result specifically includes: and planning a discrete point track meeting the collision avoidance constraint condition of the target model function by adopting a mixed A-x algorithm, and acquiring an initial value selection result by adopting the discrete point track as an initial value of the target model function.
In this embodiment, a hybrid a-x algorithm is used to plan a discrete point trajectory satisfying a collision avoidance constraint condition as an initial value of a target model function. Namely: in this embodiment, a discrete trajectory is planned by selecting the hybrid a-x algorithm, but each trajectory point satisfies the collision avoidance constraint condition of the target model function, and thus the discrete trajectory planned by the hybrid a-x algorithm is used as an initial value of the target model function.
In any of the above technical solutions, selecting a result according to the initial value, and solving the target model function to obtain a parking path planning result of the vehicle, specifically includes: and solving the target model function by adopting an IPOPT solver.
In this embodiment, an open source solver IPOPT for solving the convex optimization problem is used to solve the objective function, and a smooth natural path which is free of collision and satisfies the motion constraint is planned finally through the IPOPT numerical optimization solver.
To achieve the second object of the present invention, embodiments of the present invention provide a vehicle, which performs parking path planning by using the vehicle parking path planning method according to any of the embodiments of the present invention.
The vehicle according to the embodiment of the present invention performs parking path planning by using the vehicle parking path planning method according to any embodiment of the present invention, so that the vehicle parking path planning method according to any embodiment of the present invention has all the beneficial effects, and details are not repeated herein.
To achieve the third object of the present invention, an embodiment of the present invention provides a vehicle parking path planning apparatus including: a memory storing a computer program; a processor executing a computer program; when the processor executes the computer program, the method for planning the parking path of the vehicle according to any embodiment of the invention is implemented to plan the parking path.
The vehicle parking path planning device of the embodiment performs parking path planning by using the vehicle parking path planning method according to any embodiment of the present invention, so that the vehicle parking path planning device has all the beneficial effects of the vehicle parking path planning method according to any embodiment of the present invention, and details are not repeated herein.
To achieve the fourth object of the present invention, an embodiment of the present invention provides a computer-readable storage medium including: the computer-readable storage medium stores a computer program that, when executed, implements the steps of performing a parking path planning by a vehicle parking path planning method according to any one of the embodiments of the present invention.
The computer-readable storage medium of this embodiment performs parking path planning by using the vehicle parking path planning method according to any embodiment of the present invention, so that the computer-readable storage medium has all the beneficial effects of the vehicle parking path planning method according to any embodiment of the present invention, and details thereof are not repeated herein.
Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention.
Drawings
The above and/or additional aspects and advantages of the present invention will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings of which:
FIG. 1 is a flow chart of a first step of a vehicle parking path planning method in accordance with one embodiment of the present invention;
FIG. 2 is a diagram of a single sport vehicle model of a vehicle in accordance with one embodiment of the present invention;
fig. 3 is a schematic diagram showing the components of the vehicle parking path planning apparatus according to an embodiment of the present invention;
FIG. 4 is a flowchart of a second step of a vehicle parking path planning method in accordance with one embodiment of the present invention;
FIG. 5 is a simulation diagram of a vehicle parking path planning method in the absence of obstacles in accordance with one embodiment of the present invention;
fig. 6 is a simulation diagram of the vehicle parking path planning method in the case of an obstacle according to an embodiment of the present invention.
Wherein, the correspondence between the reference numbers and the component names in fig. 2 and 3 is:
100: vehicle, 102: front wheel, 104: rear wheel, 200: vehicle parking path planning device, 210: memory, 220: a processor.
Detailed Description
In order that the above objects, features and advantages of the present invention can be more clearly understood, a more particular description of the invention will be rendered by reference to the appended drawings. It should be noted that the embodiments and features of the embodiments of the present application may be combined with each other without conflict.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention, however, the present invention may be practiced in other ways than those specifically described herein, and therefore the scope of the present invention is not limited by the specific embodiments disclosed below.
A vehicle parking path planning method, a vehicle parking path planning apparatus 200, a vehicle 100, and a computer-readable storage medium according to some embodiments of the present invention are described below with reference to fig. 1 to 6.
Path planning is one of the important technologies in vehicle motion planning, and path planning algorithms commonly used in the related art are generally classified into 4 types, namely a geometric method, a graph-based search method, a sampling-based search method and a numerical optimization solution. The geometric method is widely applied to autonomous parking, is mainly used for obtaining information of vehicles, obstacles and surrounding environments, and plans a collision-free track by using a linear, curve or circular arc method. Although the imperfection constraint of the vehicle is met, the environmental requirement for parking is high, and generally, each geometric method is only suitable for one parking form, so that certain limitation exists. The graph-based search algorithm and the sampling-based search algorithm mainly typically represent various modified search algorithms mainly based on an a-x algorithm and a fast random search tree algorithm (abbreviated as RRT). For example, in the related art, there is a technical scheme of a self-adaptive rescue trajectory planning method based on a numerical optimization algorithm, which solves a trajectory planning problem by using a numerical optimization method, performs nonlinear programming (NLP for short) by using a numerical optimization algorithm with the aim of minimizing the working time of a thrust normal flight segment as an optimization target, and obtains an optimal solution.
However, the vehicle parking path planning method based on the numerical optimization algorithm in the related art is at the cost of increasing the calculation time, and the obtained trajectory is not an optimal solution because the numerical optimization algorithm has randomness. The numerical optimization solution is to construct an optimal objective function for the path planning problem and solve the optimal objective function through numerical optimization. The difficulty with numerical optimization algorithms is that the non-convex problem of collision avoidance constraints, which usually occurs in the form of integer variables, makes the resulting optimization problem computationally difficult to solve. Meanwhile, due to non-integrity constraint and non-convexity of a barrier-free space, the solving precision of the method depends on the selection of an initial value of a solver.
In view of the foregoing, embodiments of the present invention provide the following vehicle parking path planning method, vehicle parking path planning apparatus 200, vehicle 100, and computer-readable storage medium to provide a technical solution capable of obtaining a vehicle parking path that is smooth and satisfies motion constraints.
Example 1:
as shown in fig. 1, the present embodiment provides a vehicle parking path planning method, including:
step S102, constructing a kinematic model of the vehicle according to the size parameters and the motion state of the vehicle;
step S104, constructing an obstacle model of the vehicle according to the obstacle distribution condition around the vehicle, and acquiring a collision avoidance constraint condition of the vehicle and the obstacle;
step S106, reconstructing the collision avoidance constraint condition to ensure that the collision avoidance constraint condition has continuity;
step S108, representing the kinematic model and constructing a target model function;
step S110, selecting an initial value of the target model function, and acquiring an initial value selection result;
and step S112, solving the target model function according to the initial value selection result to obtain a parking path planning result of the vehicle.
Specifically, the embodiment provides an autonomous parking path planning method based on numerical optimization, and the method includes firstly constructing a collision avoidance constraint condition and an objective model function, converting an autonomous parking path planning problem into a non-convex optimization problem, and completely describing the environment of a general parking scene.
In order to be able to plan a path in a particularly compact parking environment, with respect to non-convexity and non-differentiability of collision avoidance, the present embodiment further proposes to reconstruct a collision avoidance constraint by introducing an auxiliary decision variable, and to introduce a penetration function into the collision avoidance constraint equation. Meanwhile, for the non-convexity of the target model function, the initial path is generated by using a hybrid a-x algorithm and initialized, and finally, a numerical solver based on a gradient descent method is used for solving.
The path obtained by the vehicle parking path planning method of the embodiment does not need to be optimized, and the requirement of autonomous parking is completely met. Simulation experiment results show that: the target model function of the vehicle parking path planning method of the embodiment takes the initial path drawn by the mixed a-x algorithm as an initial solution, and can generate a path which meets the vehicle kinematics model, is collision-free and smooth. In addition, the trajectory generated by the vehicle parking path planning method of the present embodiment is smoother and easier to track.
Example 2:
the present embodiment provides a vehicle parking path planning method, and in addition to the technical features of the above-described embodiments, the present embodiment further includes the following technical features.
In constructing a kinematic model of the vehicle based on the dimensional parameters and the kinematic state of the vehicle, the kinematic model includes:
Figure RE-RE-GDA0002614535890000081
wherein a two-dimensional plane rectangular coordinate system is constructed based on the central axis of the vehicle as the X axis, (X, Y) is the central position coordinate between two rear wheels of the vehicle,
Figure RE-RE-GDA0002614535890000082
is the angle between the front wheels of the vehicle and the X-axis, v is the speed of the vehicle in the direction of the central axis, is the steering angle of the vehicle, L is the distance between the front and rear wheels of the vehicle, and α is the acceleration of the vehicle.
Specifically, the vehicle parking path planning method of the present embodiment is applied to the vehicle 100 shown in fig. 2. In which a pair of front wheels 102 is provided at the front of the vehicle 100 and a pair of rear wheels 104 is provided at the rear. Where (X, Y) is the rear axle center of the vehicle 100, i.e.: the center position between the two rear wheels 104 is a coordinate in a two-dimensional plane orthogonal coordinate system. As shown in figure 2 of the drawings, in which,
Figure RE-RE-GDA0002614535890000083
v is the longitudinal speed of the vehicle 100, i.e. the speed of the vehicle 100 in the direction of the central axis, L is the distance between the centre point of the front wheels 102 and the centre point of the rear wheels 104 of the vehicle 100, and α is the steering angle and acceleration of the vehicle 100, respectively.
In the path planning process of the automatic driving, the planned path needs to satisfy the kinematic non-integrity constraint of the vehicle 100, so that the present embodiment adopts a vehicle dynamics single vehicle model to perform modeling, and the state of the vehicle 100 is represented by the above model formula in a low speed state.
Example 3:
the present embodiment provides a vehicle parking path planning method, and in addition to the technical features of the above-described embodiments, the present embodiment further includes the following technical features.
In the process of constructing an obstacle model of a vehicle according to the distribution condition of obstacles around the vehicle and acquiring the collision avoidance constraint condition of the vehicle and the obstacles,
the obstacle model includes:
O(m)＝{A(m)y≤b(m)|y∈R2}；
wherein A is(m)And b(m)Respectively representing known conditions, O, relating to the distribution of said obstacles(m)Representing the multicellular shape of the obstacle, R representing the real number domain, R2A real number domain representing a two-dimensional space, y representing a y coordinate point in a coordinate point function (x, y) in the real number domain;
the collision avoidance constraints include:
Figure RE-RE-GDA0002614535890000084
wherein E (x) represents the multicellular shape of the vehicle, O(m)Represents the shape of multiple cells of an obstacle in which E (x) and O are in the collision avoidance constraint(m)The intersection of (a) and (b) indicates that the vehicle does not collide with the obstacle, M indicates the number of obstacles, and M indicates the current mth obstacle.
Specifically, assuming that the obstacle model is represented by a matrix, in the present embodiment, the positions or spaces occupied by the obstacles can be represented as O, respectively(m)＝{A(m)y≤b(m)|y∈R2}. The above formula is applicable to most obstacles, and may be approximated as a union of polyhedrons.
Assuming that the vehicle 100 reaches the final state from the initial state through steering and translation, and all obstacles are to be avoided during the movement, the present embodiment expresses the collision avoidance constraint of the vehicle 100 with the obstacles as the collision avoidance constraint
Figure RE-RE-GDA0002614535890000094
(m＝1,2,…,M)。
Example 4:
the present embodiment provides a vehicle parking path planning method, and in addition to the technical features of the above-described embodiments, the present embodiment further includes the following technical features.
In reconstructing the collision avoidance constraints so that the collision avoidance constraints are continuous, the reconstructed collision avoidance constraints include:
sd(E(x),O(m))：＝dist(E(x),O(m))-pen(E(x),O(m))；
where dist is a function of the distance between the vehicle and the obstacle, pen is a function of the penetration between the vehicle and the obstacle, sd (E (x), O(m)) Represents the calculation of E (x) ∩ O(m)To determine whether the vehicle collides with an obstacle, sd (E (x), O(m)) If the result of (2) is greater than 0, E (x) ∩ O is determined(m)Is equal to
Figure RE-RE-GDA0002614535890000091
Obtaining sd (E (x), O(m)) If the result of (3) is less than or equal to 0, E (x) ∩ O is determined(m)Is not equal to
Figure RE-RE-GDA0002614535890000092
Specifically, the collision avoidance constraint condition
Figure RE-RE-GDA0002614535890000093
(M-1, 2, …, M) is generally non-convex and non-differentiable from a mathematical point of view, and it is therefore difficult to use the previous constraints in optimizing the algorithm. In view of this, the present embodiment reconstructs the collision avoidance constraint based on the concept of the symbolic distance function in the above manner to ensure the continuity thereof.
Example 5:
the present embodiment provides a vehicle parking path planning method, and in addition to the technical features of the above-described embodiments, the present embodiment further includes the following technical features.
In representing the kinematic model, constructing the objective model function,
the representation of the kinematic model includes:
xk+1＝xk+τf(xk+0.5τf(xk,uk),uk)；
where τ is the sampling time and τ is greater than 0, xk+1Is the state variable, x, of the vehicle at time k +1kIs the state variable of the vehicle at time k, ukThe control variable of the vehicle at the moment k; f (x)k+ 0.5τf(xk,uk),uk) Representing a motion state of the vehicle, the motion state including a center position coordinate (x, y) between two rear wheels of the vehicle, an angle between a front wheel of the vehicle and an x-axis
Figure RE-RE-GDA0002614535890000101
And a speed v of the vehicle in the center axis direction;
the objective model function includes:
Figure RE-RE-GDA0002614535890000102
wherein x issRepresenting the starting state of said vehicle, xFRepresenting the final state of the vehicle, R (x)k) And t (x)k) Respectively representing a rotation matrix and a translation matrix of said vehicle, G and G representing known matrices associated with a starting state of said vehicle, λ and μ being auxiliary decision variables,
Figure RE-RE-GDA0002614535890000103
and
Figure RE-RE-GDA0002614535890000104
is an obstacle O at time k(m)Of a decision variable, τ>0 is the sampling time, xkAnd ukFor the state variables and control variables, Q and Q, of the vehicle at time kΔFor a semi-positive weighting matrix, Deltau, of said vehicle with respect to a control quantityk＝(uk-uk-1) K is the weight that trades off the minimum time, h (x)k，uk) The state constraint of the vehicle is less than or equal to 0,
Figure RE-RE-GDA0002614535890000105
μ≥0：-gTμ+(At(x)-b)Tλ>d，GTμ+R(x)TATλ＝0， ||ATλ | | ═ 1 is the reconstructed collision avoidance constraint.
Specifically, the present embodiment represents the kinematic model of the vehicle 100 as x according to the second-order Runge-Kutta methodk+1＝xk+τf(xk+0.5τf(xk,uk),uk) And further constructing the target model function so that the open source solver IPOPT can solve the target model function.
Example 6
The present embodiment provides a vehicle parking path planning method, and in addition to the technical features of the above-described embodiments, the present embodiment further includes the following technical features.
Selecting an initial value of the target model function to obtain an initial value selection result, which specifically comprises the following steps: and planning a discrete point track meeting the collision avoidance constraint condition of the target model function by adopting a mixed A-x algorithm, and acquiring an initial value selection result by adopting the discrete point track as an initial value of the target model function.
Based on the non-convexity of the objective model function, the present embodiment needs to select an appropriate initial value in the solving process. In particular, the key of the numerical optimization problem is to select a proper initial value, otherwise, the algorithm may diverge, and the time and the precision of the solution are related to the selection of the initial value. When the selected initial value needs to satisfy all the constraint conditions in the target model function, the solved solution is the most accurate. Therefore, it is important to select the initial value in the process of solving the objective function.
Therefore, in the embodiment, a hybrid a-x algorithm is adopted to plan the discrete point trajectory satisfying the collision avoidance constraint condition as the initial value of the target model function. Namely: in this embodiment, a discrete trajectory is planned by selecting the hybrid a-x algorithm, but each trajectory point satisfies the collision avoidance constraint condition of the target model function, and thus the discrete trajectory planned by the hybrid a-x algorithm is used as an initial value of the target model function.
Example 7:
the present embodiment provides a vehicle parking path planning method, and in addition to the technical features of the above-described embodiments, the present embodiment further includes the following technical features.
According to the initial value selection result, solving the target model function to obtain a parking path planning result of the vehicle, and specifically comprising the following steps: and solving the target model function by adopting an IPOPT solver.
In this embodiment, an open source solver IPOPT for solving the convex optimization problem is used to solve the objective function, and a smooth natural path which is free of collision and satisfies the motion constraint is planned finally through the IPOPT numerical optimization solver.
Example 8:
as shown in fig. 2, the present embodiment provides a vehicle 100, which performs parking path planning by using the vehicle parking path planning method according to any embodiment of the present invention.
The vehicle 100 of the embodiment performs parking path planning by using the vehicle parking path planning method according to any embodiment of the present invention, so that the vehicle parking path planning method according to any embodiment of the present invention has all the beneficial effects, and details are not repeated herein.
Example 9:
as shown in fig. 3, the present embodiment provides a vehicle parking path planning apparatus 200 including: a memory 210 storing a computer program; a processor 220 executing a computer program; wherein the processor 220, when executing the computer program, implements the steps of the vehicle parking path planning method according to any of the embodiments of the present invention.
The vehicle parking path planning apparatus 200 of the present embodiment performs parking path planning by using the vehicle parking path planning method according to any embodiment of the present invention, so that it has all the beneficial effects of the vehicle parking path planning method according to any embodiment of the present invention, and details thereof are not repeated herein.
Example 10:
the present embodiments provide a computer-readable storage medium, comprising: the computer-readable storage medium stores a computer program that, when executed, implements the steps of a vehicle parking path planning method according to any one of the embodiments of the present invention.
The computer-readable storage medium of this embodiment performs parking path planning by using the vehicle parking path planning method according to any embodiment of the present invention, so that the computer-readable storage medium has all the beneficial effects of the vehicle parking path planning method according to any embodiment of the present invention, and details thereof are not repeated herein.
DETAILED DESCRIPTION OF EMBODIMENT (S) OF INVENTION
As shown in fig. 4, the present embodiment provides a vehicle parking path planning method, which includes the following steps:
step S402, constructing an obstacle model;
step S404, reconstructing collision avoidance constraints;
step S406, constructing a target model function;
step S408, mixing A to calculate an initial value;
step S410, solving an IPOPT numerical optimization solver;
step S412, judging whether no collision exists and motion constraint is met;
if yes, step S408 is executed again, and if no, the process is terminated.
First, the present embodiment establishes a kinematic model of the vehicle. Specifically, in the process of planning a path for automatic driving, the planned path needs to satisfy kinematic non-integrity constraints of a vehicle, in this embodiment, a vehicle dynamics single-vehicle model is used for modeling, and in a low-speed state, a state of the vehicle may be represented as:
Figure RE-RE-GDA0002614535890000121
where (X, Y) is the rear axle center of the vehicle, v is the longitudinal velocity, is the steering angle, and a is the acceleration.
Further, the present embodiment performs obstacle model construction. Wherein, assuming that the obstacle model is represented by a matrix, the situation occupied by the obstacle can be represented as:
O(m)＝{A(m)y≤b(m)|y∈R2}；
wherein A is(m)And b(m)Respectively representing known conditions, O, relating to the distribution of said obstacles(m)Representing the multicellular shape of the obstacle, R representing the real number domain, R2Represents a real number domain of a two-dimensional space, and y represents a y-coordinate point in a coordinate point function (x, y) in the real number domain.
It should be noted that the above formula is applicable to most obstacles, and may be approximated as a union of polyhedrons. Assuming that the vehicle reaches the final state from the initial state by turning and translating, and all obstacles are to be avoided during the movement, the collision avoidance constraint of the vehicle with the obstacles may be expressed as:
Figure RE-RE-GDA0002614535890000131
wherein E (x) represents the multicellular shape of the vehicle, O(m)Represents the shape of multiple cells of an obstacle in which E (x) and O are in the collision avoidance constraint(m)The intersection of (a) and (b) indicates that the vehicle does not collide with the obstacle, M indicates the number of obstacles, and M indicates the current mth obstacle.
Mathematically, the above equation is generally non-convex and non-differentiable, so it is difficult to use the previous constraints in optimizing the algorithm. For this reason, the present embodiment needs to reconstruct it to ensure its continuity.
Subsequently, the present embodiment implements the step of reconstructing the collision avoidance constraint. Specifically, the present embodiment reconstructs the collision avoidance constraint condition based on the concept of the symbol distance function, where the collision avoidance constraint condition is:
sd(E(x),O(m))：＝dist(E(x),O(m))-pen(E(x),O(m))；
where dist and pen represent the distance function and penetration function of the vehicle from the obstacle, sd (E (x), O, respectively(m)) Represents the calculation of E (x) ∩ O(m)As a result of (A)To determine whether the vehicle collides with an obstacle or not, and to obtain sd (E (x), O(m)) If the result of (2) is greater than 0, E (x) ∩ O is determined(m)Is equal to
Figure RE-RE-GDA0002614535890000132
Obtaining sd (E (x), O(m)) If the result of (3) is less than or equal to 0, E (x) ∩ O is determined(m)Is not equal to
Figure RE-RE-GDA0002614535890000133
Further, the present embodiment constructs an objective model function, that is: according to the second-order Runge-Kutta method, the vehicle kinematics model is represented as:
xk+1＝xk+τf(xk+0.5τf(xk,uk),uk)；
in the above formula, τ is the sampling time and is greater than 0, xk+1Is the state variable, x, of the vehicle at time k +1k,ukState variables and control variables of the vehicle at the moment k; f (x)k+0.5τf(xk,uk),uk) Representing a motion state of the vehicle, the motion state including a center position coordinate (x, y) between two rear wheels of the vehicle, an angle between a front wheel of the vehicle and an x-axis
Figure RE-RE-GDA0002614535890000141
And a speed v of the vehicle in the center axis direction.
The objective model function of this embodiment is:
Figure RE-RE-GDA0002614535890000142
wherein x issRepresenting the starting state of said vehicle, xFRepresenting the final state of the vehicle, R (x)k) And t (x)k) Respectively representing a rotation matrix and a translation matrix of said vehicle, G and G representing known matrices associated with a starting state of said vehicle, λ and μ being auxiliary decision variables,
Figure RE-RE-GDA0002614535890000143
and
Figure RE-RE-GDA0002614535890000144
is an obstacle O at time k(m)Of a decision variable, τ>0 is the sampling time, xkAnd ukFor the state variables and control variables, Q and Q, of the vehicle at time kΔFor a semi-positive weighting matrix, Deltau, of said vehicle with respect to a control quantityk＝(uk-uk-1) K is the weight that trades off the minimum time, h (x)k，uk) The state constraint of the vehicle is less than or equal to 0,
Figure RE-RE-GDA0002614535890000145
Figure RE-RE-GDA0002614535890000146
μ≥0：-gTμ+(At(x)-b)Tλ>d，GTμ+R(x)TATλ＝0，||ATλ | | ═ 1 is the reconstructed collision avoidance constraint.
Further, the present embodiment performs initial value selection. Specifically, based on the non-convexity of the objective model function, a proper initial value needs to be selected in the solving process, the key of the numerical optimization problem is to select the proper initial value, otherwise, the algorithm may be divergent, and the solving time and precision are related to the selection of the initial value. When the selected initial value needs to satisfy all the constraint conditions in the target model function, the solved solution is the most accurate. Therefore, the initial value selection in the process of solving the target model function is very important. In this embodiment, a discrete point trajectory satisfying the constraint condition of the objective function is planned by using a hybrid a-x algorithm, and is used as an initial value of the objective function. Therefore, in the embodiment, a discrete track is planned by selecting the hybrid a-x algorithm, but each track point meets the constraint condition of the target model function. And taking the discrete track planned by the mixed A-x algorithm as an initial value of the target model function.
Finally, the present embodiment solves the objective model function. Namely: in this embodiment, an open source solver IPOPT for solving a convex optimization problem is used to solve a target model function, and a smooth natural path which is free of collision and satisfies motion constraints is finally planned through the IPOPT numerical optimization solver.
The implementation converts the autonomous parking path planning problem into a non-convex optimization problem, reconstructs collision avoidance constraint conditions, constructs an optimal target model function for parking path planning, plans an initial path by adopting a mixed A-star algorithm as an initial value, and solves a path which meets a vehicle kinematics model, is collision-free and smooth by calling an IPOPT numerical optimization solver. The simulation results of the vehicle parking path planning described above in this embodiment are shown in fig. 5 and 6. According to the method and the device, a path which meets a vehicle kinematics model, is collision-free and smooth can be generated under the problem of planning of the autonomous parking path, and compared with a traditional graph-based search algorithm, the generated track is smoother and easier to track.
In addition, it should be noted that the autonomous parking path planning method based on the B-spline theory can also achieve the object of the embodiment of the present invention, and by establishing an obstacle avoidance constraint function, establishing a constraint function of a vehicle steering angle and an angular velocity, and establishing a plurality of parking path objective functions with nonlinear constraints with the vehicle pose information of the parking end point position, a parking path with continuous curvature and slow change can be obtained, and the obstacle constraints and other constraints can be satisfied.
In summary, the embodiment of the invention has the following beneficial effects: the embodiment of the invention can realize the planning of the autonomous parking path and generate a smooth parking path which meets the vehicle kinematics model and has no collision.
In the present invention, the terms "first", "second", and "third" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance; the term "plurality" means two or more unless expressly limited otherwise. The terms "mounted," "connected," "fixed," and the like are to be construed broadly, and for example, "connected" may be a fixed connection, a removable connection, or an integral connection; "coupled" may be direct or indirect through an intermediary. The specific meanings of the above terms in the present invention can be understood by those skilled in the art according to specific situations.
In the description of the present invention, it is to be understood that the terms "upper", "lower", "left", "right", "front", "rear", and the like indicate orientations or positional relationships based on those shown in the drawings, and are only for convenience of description and simplification of description, but do not indicate or imply that the referred device or unit must have a specific direction, be constructed in a specific orientation, and be operated, and thus, should not be construed as limiting the present invention.
In the description herein, the description of the terms "one embodiment," "some embodiments," "specific embodiments," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
The above is only a preferred embodiment of the present invention, and is not intended to limit the present invention, and various modifications and changes will occur to those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims (10)

1. A vehicle parking path planning method, comprising:
constructing a kinematic model of the vehicle according to the size parameters and the motion state of the vehicle;
according to the distribution condition of obstacles around the vehicle, constructing an obstacle model of the vehicle, and acquiring a collision avoidance constraint condition of the vehicle and the obstacles;
reconstructing the collision avoidance constraints such that the collision avoidance constraints are continuous;
representing the kinematic model and constructing a target model function;
selecting an initial value of the target model function to obtain an initial value selection result;
and solving the target model function according to the initial value selection result to obtain a parking path planning result of the vehicle.
2. The vehicle parking path planning method according to claim 1, wherein in the constructing of the kinematic model of the vehicle according to the size parameter and the motion state of the vehicle, the kinematic model includes:
Figure RE-FDA0002614535880000011
wherein a two-dimensional plane rectangular coordinate system is constructed based on a central axis of the vehicle as an x-axis, (x, y) being a central position coordinate between two rear wheels of the vehicle,
Figure RE-FDA0002614535880000012
is the angle between the front wheels of the vehicle and the x-axis, v is the speed of the vehicle in the direction of the central axis, is the steering angle of the vehicle, L is the distance between the front and rear wheels of the vehicle, α is the acceleration of the vehicle.
3. The vehicle parking path planning method according to claim 2, wherein in the obtaining of the collision avoidance constraint condition of the vehicle and the obstacle, constructing an obstacle model of the vehicle based on the obstacle distribution situation around the vehicle,
the obstacle model includes:
O(m)＝{A(m)y≤b(m)|y∈R2}；
wherein A is(m)And b(m)Respectively representing known conditions, O, relating to the distribution of said obstacles(m)Representing the multicellular shape of the obstacle, R representing the real number domain, R2A real number domain representing a two-dimensional space, y representing a y coordinate point in a coordinate point function (x, y) in the real number domain;
the collision avoidance constraints include:
Figure RE-FDA0002614535880000021
wherein E (x) represents the multicellular shape of the vehicle, O(m)Represents the shape of multiple cells of an obstacle in which E (x) and O are in the collision avoidance constraint(m)The intersection of (a) and (b) indicates that the vehicle does not collide with the obstacle, M indicates the number of obstacles, and M indicates the current mth obstacle.
4. The vehicle parking path planning method according to claim 3, wherein in the reconstructing the collision avoidance constraints so that the collision avoidance constraints have continuity, the reconstructed collision avoidance constraints include:
sd(E(x),O(m))：＝dist(E(x),O(m))-pen(E(x),O(m))；
where dist is a function of the distance between the vehicle and the obstacle, pen is a function of the penetration between the vehicle and the obstacle, sd (E (x), O(m)) Represents the calculation of E (x) ∩ O(m)To determine whether the vehicle collides with an obstacle, sd (E (x), O(m)) If the result of (2) is greater than 0, E (x) ∩ O is determined(m)Is equal to
Figure RE-FDA0002614535880000023
Obtaining sd (E (x), O(m)) If the result of (3) is less than or equal to 0, E (x) ∩ O is determined(m)Is not equal to
Figure RE-FDA0002614535880000024
5. The vehicle parking path planning method according to any one of claims 1 to 4, wherein, in the representing the kinematic model, constructing an object model function,
the representation of the kinematic model includes:
xk+1＝xk+τf(xk+0.5τf(xk,uk),uk)；
where τ is the sampling time and τ is greater than 0, xk+1Is the state variable, x, of the vehicle at time k +1kIs the state variable of the vehicle at time k, ukThe control variable of the vehicle at the moment k; f (x)k+0.5τf(xk,uk),uk) Representing a motion state of the vehicle, the motion state including a center position coordinate (x, y) between two rear wheels of the vehicle, an angle between a front wheel of the vehicle and an x-axis
Figure RE-FDA0002614535880000025
And a speed v of the vehicle in the center axis direction;
the objective model function includes:
Figure RE-FDA0002614535880000022
Figure RE-FDA0002614535880000031
wherein x issRepresenting the starting state of said vehicle, xFRepresenting the final state of the vehicle, R (x)k) And t (x)k) Respectively representing a rotation matrix and a translation matrix of said vehicle, G and G representing known matrices associated with a starting state of said vehicle, λ and μ being auxiliary decision variables,
Figure RE-FDA0002614535880000032
and
Figure RE-FDA0002614535880000033
is an obstacle O at time k(m)Of a decision variable, τ>0 is the sampling time, xkAnd ukFor the state variables and control variables, Q and Q, of the vehicle at time kΔFor a semi-positive weighting matrix, Deltau, of said vehicle with respect to a control quantityk＝(uk-uk-1) K is the weight that trades off the minimum time, h (x)k，uk) ≦ 0 is the state constraint for the vehicle,
Figure RE-FDA0002614535880000034
μ≥0：-gTμ+(At(x)-b)Tλ>d，GTμ+R(x)TATλ＝0，||ATλ | | ═ 1 is the reconstructed collision avoidance constraint.
6. The vehicle parking path planning method according to any one of claims 1 to 4, wherein the initial value selection of the target model function to obtain an initial value selection result specifically includes:
and planning a discrete point track meeting the collision avoidance constraint condition of the target model function by adopting a mixed A-x algorithm, and acquiring an initial value selection result by adopting the discrete point track as an initial value of the target model function.
7. The vehicle parking path planning method according to any one of claims 1 to 4, wherein the obtaining of the parking path planning result of the vehicle by solving the objective model function according to the initial value selection result specifically includes:
and solving the target model function by adopting an IPOPT solver.
8. A vehicle characterized in that a parking path is planned using the vehicle parking path planning method according to any one of claims 1 to 7.
9. A vehicle parking path planning apparatus characterized by comprising:
a memory storing a computer program;
a processor executing the computer program;
wherein the processor, when executing the computer program, implements the steps of the vehicle parking path planning method according to any one of claims 1 to 7.
10. A computer-readable storage medium, comprising:
the computer-readable storage medium stores a computer program that, when executed, implements the steps of the vehicle parking path planning method according to any one of claims 1 to 7.
CN202010608892.9A 2020-06-30 2020-06-30 Vehicle parking path planning method and device, vehicle and storage medium Pending CN111746523A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010608892.9A CN111746523A (en) 2020-06-30 2020-06-30 Vehicle parking path planning method and device, vehicle and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010608892.9A CN111746523A (en) 2020-06-30 2020-06-30 Vehicle parking path planning method and device, vehicle and storage medium

Publications (1)

Publication Number Publication Date
CN111746523A true CN111746523A (en) 2020-10-09

Family

ID=72678100

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010608892.9A Pending CN111746523A (en) 2020-06-30 2020-06-30 Vehicle parking path planning method and device, vehicle and storage medium

Country Status (1)

Country Link
CN (1) CN111746523A (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113335312A (en) * 2021-08-06 2021-09-03 新石器慧通（北京）科技有限公司 Obstacle-detouring driving method and device, electronic equipment and medium
CN113799766B (en) * 2021-09-17 2023-04-18 东风汽车集团股份有限公司 Constrained parallel parking path planning method, electronic device and storage medium
WO2023206995A1 (en) * 2022-04-27 2023-11-02 广州汽车集团股份有限公司 Parking path planning method and system

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108121205A (en) * 2017-12-13 2018-06-05 深圳市航盛电子股份有限公司 A kind of paths planning method, system and medium for a variety of scenes of parking
DE102018003086A1 (en) * 2018-04-16 2018-10-25 Daimler Ag Method for automatically starting a parking space
CN110065490A (en) * 2019-04-30 2019-07-30 驭势（上海）汽车科技有限公司 Automatic parking method, system, automatic driving vehicle and storage medium
US20190329761A1 (en) * 2015-08-12 2019-10-31 Hyundai Motor Company Automatic parking system and automatic parking method
CN110806218A (en) * 2019-11-29 2020-02-18 北京京东乾石科技有限公司 Parking path planning method, device and system
CN111301409A (en) * 2020-03-11 2020-06-19 中国第一汽车股份有限公司 Parking path planning method and device, vehicle and storage medium

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20190329761A1 (en) * 2015-08-12 2019-10-31 Hyundai Motor Company Automatic parking system and automatic parking method
CN108121205A (en) * 2017-12-13 2018-06-05 深圳市航盛电子股份有限公司 A kind of paths planning method, system and medium for a variety of scenes of parking
DE102018003086A1 (en) * 2018-04-16 2018-10-25 Daimler Ag Method for automatically starting a parking space
CN110065490A (en) * 2019-04-30 2019-07-30 驭势（上海）汽车科技有限公司 Automatic parking method, system, automatic driving vehicle and storage medium
CN110806218A (en) * 2019-11-29 2020-02-18 北京京东乾石科技有限公司 Parking path planning method, device and system
CN111301409A (en) * 2020-03-11 2020-06-19 中国第一汽车股份有限公司 Parking path planning method and device, vehicle and storage medium

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
XIAOJING ZHANG ET AL: ""Autonomous Parking using Optimization-Based Collision Avoidance"", 《2018 IEEE CONFERENCE ON DECISION AND CONTROL (CDC)》 *
XIAOJING ZHANG ET AL: ""Optimization-Based Collision Avoidance"", 《IEEE TRANSACTIONS ON CONTROL SYSTEMS TECHNOLOGY》 *

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113335312A (en) * 2021-08-06 2021-09-03 新石器慧通（北京）科技有限公司 Obstacle-detouring driving method and device, electronic equipment and medium
CN113799766B (en) * 2021-09-17 2023-04-18 东风汽车集团股份有限公司 Constrained parallel parking path planning method, electronic device and storage medium
WO2023206995A1 (en) * 2022-04-27 2023-11-02 广州汽车集团股份有限公司 Parking path planning method and system

Similar Documents

Publication Publication Date Title
CN111746523A (en) Vehicle parking path planning method and device, vehicle and storage medium
CN109885891B (en) Intelligent vehicle GPU parallel acceleration trajectory planning method
Ali et al. Path planning of mobile robot with improved ant colony algorithm and MDP to produce smooth trajectory in grid-based environment
Fu et al. Phase angle-encoded and quantum-behaved particle swarm optimization applied to three-dimensional route planning for UAV
Hasircioglu et al. 3-D path planning for the navigation of unmanned aerial vehicles by using evolutionary algorithms
Xue et al. Trajectory planning for autonomous mobile robot using a hybrid improved QPSO algorithm
CN109434831B (en) Robot operation method and device, robot, electronic device and readable medium
Rubies-Royo et al. A classification-based approach for approximate reachability
CN110181508B (en) Three-dimensional route planning method and system for underwater robot
CN103984230B (en) A kind of space manipulator pedestal zero disturbs optimal control method
CN110543727B (en) Improved particle swarm algorithm-based omnidirectional mobile intelligent wheelchair robot parameter identification method
Bouton et al. Utility decomposition with deep corrections for scalable planning under uncertainty
CN116700327A (en) Unmanned aerial vehicle track planning method based on continuous action dominant function learning
Marchese A directional diffusion algorithm on cellular automata for robot path-planning
Hu Research on robot fuzzy neural network motion system based on artificial intelligence
Ouyang et al. Virtual velocity vector-based offline collision-free path planning of industrial robotic manipulator
Li et al. Research on the agricultural machinery path tracking method based on deep reinforcement learning
CN116009558A (en) Mobile robot path planning method combined with kinematic constraint
Chen et al. From topological map to local cognitive map: a new opportunity of local path planning
Hua et al. A Novel Learning-Based Trajectory Generation Strategy for a Quadrotor
Hamdan et al. Study of Path Planning Methods in Two-Dimensional Mapped Environments
CN114610047B (en) QMM-MPC underwater robot vision docking control method for online depth estimation
CN113296536B (en) Unmanned aerial vehicle three-dimensional obstacle avoidance algorithm based on A-star and convex optimization algorithm
Li et al. Relaxation labeling using augmented Lagrange-Hopfield method
Aguilar et al. PSO trajectory planner using kinematic controllers that ensure smooth differential robot velocities

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
RJ01 Rejection of invention patent application after publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20201009