Mobile robot real-time path planning method and related device

Abstract

The invention provides a real-time path planning method for a mobile robot and a related device, and belongs to the technical field of robots. According to the method, state variables and control variables in a dynamic model of the mobile robot are discretized in a state space, and a discrete state space is obtained. Generalized cell mapping is established on the discrete state space for each state variable and control variable of the kinetic model, the constraint condition and the target function, and the cell mapping relation of all controllable cells in the discrete state space based on the transition probability matrix is obtained. And based on the obtained cell mapping relation of all the controllable cells based on the transition probability matrix in the discrete state space, performing optimal search according to a target function of the optimal path planning model of the mobile robot to obtain an optimal result. And performing real-time path planning on the mobile robot according to the obtained optimal result to obtain a real-time path planning result. The problem that the accuracy of real-time path planning of the mobile robot is not high is solved.

Description

Technical Field

[0001] The present invention belongs to the technical field of robots, and particularly relates to a real-time path planning method and related device for a mobile robot. Background Art

[0002] In a complex environment, when a mobile robot performs logistics distribution and personnel guidance tasks, it must rely on a safe and efficient path. Therefore, it is crucial to improve the path planning performance of the mobile robot. Thus, an efficient path planning algorithm for a mobile robot suitable for a complex environment is essential.

[0003] During navigation, the navigation system needs to plan the specific travel route of the robot from the current position to the target position, while considering avoiding obstacles and meeting specific constraints, such as the shortest path or the minimum pathfinding cost, etc. In path planning with a map, the path planning algorithm requires the robot to first build a global map before performing path planning. Even in path planning with a map, there will still be various inevitable unexpected situations. Traditional path planning algorithms are relatively difficult to model dynamic complex environments, with slow algorithm convergence speed and are difficult to apply to unknown environments.

[0004] Currently, the path planning methods for mobile robots are in a stage of continuous improvement and perfection. Although the geometric line construction method can quickly generate an optimal path, due to the discontinuity of the trajectory, the accuracy is significantly affected. The graph search method based on the Dijkstra algorithm not only takes a long time to search but also lacks effectiveness. Random sampling methods (such as probabilistic roadmap and rapidly-exploring random tree), although they can efficiently find the optimal path in a vast environment, cannot effectively solve the path planning problem due to insufficient sampling points in a local narrow environment.

[0005] With the rapid development of artificial intelligence technology and computer hardware, various intelligent algorithms, such as neural networks, fuzzy control, machine learning, genetic algorithms, and particle swarm optimization, have been widely applied to the path planning of mobile robots and have achieved good results. However, the complex calculation processes of these intelligent algorithms often cannot meet the real-time requirements, and there are also significant deficiencies in terms of accuracy, stability, and security. Summary of the Invention

[0006] The purpose of the present invention is to provide a real-time path planning method and related device for a mobile robot, which is used to solve the problem of low accuracy in the real-time path planning of mobile robots in the prior art.

[0007] To achieve the above purpose, the present invention adopts the following technical solutions: In the first aspect, the present invention provides a real-time path planning method for a mobile robot, including the following steps: Determine the observation space, state space, and action space of the mobile robot; Construct a dynamic model of the mobile robot based on its observation space, state space, and action space, and set constraint conditions, where the constraint conditions include kinematic constraints of the mobile robot, boundary conditions, and obstacle constraints during the movement of the mobile robot; Construct an optimal path planning model for the mobile robot from the initial position to the target position, where the optimal path planning model of the mobile robot takes minimizing the travel time or minimizing the travel distance as the objective function; Discretize the state variables and control variables in the dynamic model of the mobile robot in the state space to obtain a discrete state space; Establish a generalized cell mapping for each state variable and control variable of the dynamic model, the constraint conditions, and the objective function in the discrete state space to obtain the cell mapping relationship based on the transition probability matrix of all controllable cells in the discrete state space; Based on the obtained cell mapping relationship based on the transition probability matrix of all controllable cells in the discrete state space, perform an optimal search according to the objective function of the optimal path planning model of the mobile robot to obtain an optimal result; Perform real-time path planning for the mobile robot according to the obtained optimal result to obtain a real-time path planning result.

[0008] A further improvement of the present invention is that the step of discretizing the state variables and control variables in the dynamic model of the mobile robot in the state space to obtain a discrete state space specifically includes: Discretize the state variables and control variables in the dynamic model of the mobile robot in the state space into several cells to obtain a discrete state space.

[0009] A further improvement of the present invention is that the establishment of the generalized cell mapping is specifically: using GPU parallel computing technology to establish a generalized cell mapping for each state variable and control variable of the dynamic model, the constraint conditions, and the objective function in the discrete state space.

[0010] A further improvement of the present invention is that in the step of performing an optimal search according to the objective function of the optimal path planning model of the mobile robot based on the obtained cell mapping relationship based on the transition probability matrix of all controllable cells in the discrete state space to obtain an optimal result, the Bellman optimality principle is specifically used to perform a global optimal search according to the objective function of the optimal path planning model of the mobile robot to obtain a global optimal result.

[0011] A further improvement of the present invention lies in that, in the step of performing global optimal search according to the objective function of the optimal path planning model of the mobile robot by using the Bellman optimal principle to obtain the global optimal result, an obstacle detection function is also set to detect obstacles.

[0012] A further improvement of the present invention lies in that, in the step of performing global optimal search according to the objective function of the optimal path planning model of the mobile robot by using the Bellman optimal principle to obtain the global optimal result, first an obstacle detection function is set to detect obstacles, and then a subdivision method is used to perform local optimal search to obtain the local optimal result.

[0013] A further improvement of the present invention lies in that the optimal result is presented in the form of a table.

[0014] In a second aspect, the present invention provides a real-time path planning system for a mobile robot, including a space determination module, a dynamics model construction module, an optimal path planning model construction module, a discretization module, a generalized cell mapping module, an objective function optimal search module, and a path planning module; The space determination module is used to determine the observation space, state space, and action space of the mobile robot; The dynamics model construction module is used to construct the dynamics model of the mobile robot according to the observation space, state space, and action space of the mobile robot, and set constraint conditions, where the constraint conditions include the kinematic constraints of the mobile robot, boundary conditions, and obstacle constraints during the movement of the mobile robot; The optimal path planning model construction module is used to construct an optimal path planning model for the mobile robot from the initial position to the target position, and the optimal path planning model of the mobile robot takes minimizing the travel time or minimizing the travel distance as the objective function; The discretization module is used to discretize the state variables and control variables in the dynamics model of the mobile robot in the state space to obtain a discrete state space; The generalized cell mapping module is used to establish a generalized cell mapping for each state variable and control variable of the dynamics model, the constraint conditions, and the objective function in the discrete state space to obtain the cell mapping relationship based on the transition probability matrix of all controllable cells in the discrete state space; The objective function optimal search module is used to perform optimal search based on the obtained cell mapping relationship based on the transition probability matrix of all controllable cells in the discrete state space according to the objective function of the optimal path planning model of the mobile robot to obtain the optimal result; The path planning module is used to perform real-time path planning on the mobile robot according to the obtained optimal result to obtain the real-time path planning result.

[0015] In a third aspect, the present invention provides an electronic device, including a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, the steps of the real-time path planning method for a mobile robot introduced above are implemented.

[0016] In a fourth aspect, the present invention provides a storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the real-time path planning method for a mobile robot introduced above are implemented.

[0017] Compared with the prior art, the present invention has the following beneficial effects: The present invention belongs to an improved invention. Compared with the existing real-time path planning methods for mobile robots, on the one hand, the present invention establishes a generalized cell mapping for each state variable and control variable of the dynamic model, constraint conditions, and objective function in the discrete state space, and obtains the cell mapping relationship based on the transition probability matrix of all controllable cells in the discrete state space. This can not only cover the state space within the cell more comprehensively, thereby capturing the dynamic behavior of the robot more accurately. Moreover, establishing a generalized cell mapping for each state variable and control variable of the dynamic model, constraint conditions, and objective function in the discrete state space can reduce the discretization error and improve the accuracy of subsequent path planning.

[0018] Furthermore, the present invention discloses using GPU parallel computing technology to establish a generalized cell mapping for each state variable and control variable of the dynamic model, constraint conditions, and objective function in the discrete state space. The GPU parallel computing technology has powerful parallel computing capabilities, and thus can greatly improve the computing efficiency.

[0019] Furthermore, the present invention discloses using the Bellman optimality principle to perform global optimal search according to the objective function of the optimal path planning model of the mobile robot to obtain the global optimal result. The Bellman optimality principle can ensure that the found path is the global optimal path rather than a local optimal path. Moreover, the Bellman optimality principle can reduce unnecessary iteration times, and thus can reduce the computational complexity.

[0020] Furthermore, in the steps of using the Bellman optimality principle to perform global optimal search according to the objective function of the optimal path planning model of the mobile robot to obtain the global optimal result, an obstacle detection function is also set to detect obstacles, which can help avoid getting into an infeasible region when using the Bellman optimality principle, and improve the efficiency of the optimal search and the quality of the global optimal result. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 is a flowchart of the real-time path planning method for the mobile robot of the present invention; Figure 2Schematic diagram of the real-time path planning system for the mobile robot of the present invention; Figure 3 Flowchart of the real-time path planning method for the mobile robot in Embodiment 3 of the present invention; Figure 4 Scene diagram of the real-time path planning for the mobile robot in Embodiment 3 of the present invention; Figure 5 Schematic diagram of the generalized cell mapping principle in Embodiment 3 of the present invention; Figure 6 Flowchart of the global optimal search in Embodiment 3 of the present invention; Figure 7 Flowchart of the local optimal search in Embodiment 3 of the present invention; Figure 8 Schematic diagram of the structure of the electronic device of the present invention. Detailed implementation manners

[0022] To further understand the content of the present invention, the following describes the present invention in detail with reference to the accompanying drawings and specific embodiments. It should be understood that the embodiments are only for explaining the present invention rather than limiting it.

[0023] The real-time path planning method for the mobile robot proposed by the present invention establishes a generalized cell mapping for each state variable and control variable of the dynamic model, constraint conditions and objective function in the discrete state space, and obtains the cell mapping relationship based on the transition probability matrix of all controllable cells in the discrete state space. Based on the obtained cell mapping relationship of all controllable cells in the discrete state space based on the transition probability matrix, an optimal search is performed according to the objective function of the optimal path planning model of the mobile robot to obtain an optimal result. According to the obtained optimal result, the mobile robot is subjected to real-time path planning to obtain a real-time path planning result. Compared with the prior art, the present invention effectively solves the problem of low accuracy in the real-time path planning of the mobile robot in the prior art.

[0024] Embodiment 1: The flowchart of the real-time path planning method for the mobile robot of the present invention is as Figure 1 shown. The real-time path planning method for the mobile robot of the present invention includes the following steps: S1. Determine the observation space, state space and action space of the mobile robot.

[0025] S2. Construct a dynamic model of the mobile robot according to the observation space, state space and action space of the mobile robot, and set constraint conditions, where the constraint conditions include kinematic constraints of the mobile robot, boundary conditions and obstacle constraints during the movement of the mobile robot.

[0026] S3. Construct an optimal path planning model for the mobile robot from the initial position to the target position. The optimal path planning model for the mobile robot takes minimizing the travel time or minimizing the travel distance as the objective function.

[0027] S4. Discretize the state variables and control variables in the dynamic model of the mobile robot in the state space to obtain a discrete state space.

[0028] S5. Establish a generalized cell mapping for each state variable and control variable of the dynamic model, the constraint conditions, and the objective function in the discrete state space to obtain the cell mapping relationship based on the transition probability matrix for all controllable cells in the discrete state space.

[0029] S6. Based on the obtained cell mapping relationship based on the transition probability matrix for all controllable cells in the discrete state space, perform an optimal search according to the objective function of the optimal path planning model for the mobile robot to obtain the optimal result.

[0030] S7. Perform real-time path planning for the mobile robot according to the obtained optimal result to obtain the real-time path planning result.

[0031] Embodiment 2: The schematic diagram of the real-time path planning system for the mobile robot of the present invention is as Figure 2 shown. The real-time path planning system for the mobile robot of the present invention includes a space determination module, a dynamic model construction module, an optimal path planning model construction module, a discretization module, a generalized cell mapping module, an objective function optimal search module, and a path planning module.

[0032] The space determination module is used to determine the observation space, state space, and action space of the mobile robot.

[0033] The dynamic model construction module is used to construct the dynamic model of the mobile robot according to the observation space, state space, and action space of the mobile robot and set the constraint conditions. The constraint conditions include the kinematic constraints of the mobile robot, boundary conditions, and obstacle constraints during the movement of the mobile robot.

[0034] The optimal path planning model construction module is used to construct an optimal path planning model for the mobile robot from the initial position to the target position. The optimal path planning model for the mobile robot takes minimizing the travel time or minimizing the travel distance as the objective function; The discretization module is used to discretize the state variables and control variables in the dynamic model of the mobile robot in the state space to obtain a discrete state space.

[0035] The generalized cell mapping module is used to establish a generalized cell mapping for each state variable and control variable of the dynamic model, the constraint condition, and the objective function in the discrete state space, and obtain the cell mapping relationship based on the transition probability matrix for all controllable cells in the discrete state space.

[0036] The objective function optimal search module is used to perform an optimal search based on the objective function of the mobile robot optimal path planning model according to the obtained cell mapping relationship based on the transition probability matrix for all controllable cells in the discrete state space, and obtain the optimal result.

[0037] The path planning module is used to perform real-time path planning for the mobile robot according to the obtained optimal result, and obtain the real-time path planning result.

[0038] Embodiment 3: The flowchart of the real-time path planning method for the mobile robot of the present invention is as Figure 3 shown. The real-time path planning method for the mobile robot of the present invention includes the following steps: S1. Determine the observation space, state space, and action space of the mobile robot.

[0039] First, determine the observation space, state space, and action space of the mobile robot according to the mobile robot path planning scenario.

[0040] In this embodiment, the mobile robot path planning scenario in a complex dynamic environment is as Figure 4 shown. Specifically, the mobile robot path planning scenario is a dynamic pedestrian occlusion scenario in a multi-room environment, where pedestrians in the dynamic environment occlude the mobile robot dynamically, increasing the difficulty of path planning. Fix the starting point and the target point of the robot. The driving path of Robot 1 is from Room A to Room D, and the driving path of Robot 2 is from B to Room C. There are moving pedestrians walking towards each other in the corridor ( Figure 4 represented by Pedestrian A and Pedestrian B in the figure). The following takes this scenario as an example to elaborate on the optimal path planning of the mobile robot in detail: The observation space is determined based on the robot's monitoring of the environment, and the observation space is composed of a two-dimensional lidar map, pedestrian motion characteristics, and the target pose of the robot. The state space of the mobile robot is the set of all states within the driving space, including the position of the mobile robot in the lidar map and the distance between the robot and the obstacle. The action space of the mobile robot is the speed command of the differential drive robot, which is composed of the linear velocity and angular velocity of the robot at a certain moment. The passing pedestrians and obstacles such as rooms are set as constraints. The main goal of the mobile robot path planning is to enable the robot to safely and quickly avoid moving obstacles and reach the target position. In this embodiment, minimizing the travel time or minimizing the travel distance is used as the objective function. The optimal path planning of the mobile robot is a planning scenario from the initial position to the termination position (also called the target position) under the given constraints and satisfying the objective function (the minimum travel time or travel distance). The optimal path planning of the mobile robot is carried out in a local narrow environment at a low speed, and reaches the target position while meeting the requirements of safety, comfort, convenience and speed.

[0041] S2. Construct a dynamic model of the mobile robot according to the observation space, state space and action space of the mobile robot, and set constraints, including the kinematic constraints of the mobile robot, boundary conditions, and obstacle constraints during the movement of the mobile robot.

[0042] S3. Construct an optimal path planning model of the mobile robot from the initial position to the target position. The optimal path planning model of the mobile robot uses minimizing the travel time or minimizing the travel distance as the objective function.

[0043] S4. Discretize the state variables and control variables in the dynamic model of the mobile robot in the state space to obtain a discrete state space.

[0044] In this step, discretizing the state variables and control variables in the dynamic model of the mobile robot in the state space to obtain a discrete state space specifically includes: Discretize the state variables and control variables in the dynamic model of the mobile robot in the state space into several cells to obtain a discrete state space.

[0045] S5. Establish a generalized cell mapping for each state variable and control variable of the dynamic model, constraints, and objective function in the discrete state space to obtain the cell mapping relationship based on the transition probability matrix of all controllable cells in the discrete state space.

[0046] In this embodiment, when establishing the generalized cell mapping, the GPU parallel computing technology is specifically used to establish the generalized cell mapping for each state variable and control variable of the dynamic model, constraints, and objective function in the discrete state space. The following is a detailed introduction to the generalized cell mapping: In this step, the schematic diagram of the generalized cell mapping principle is as Figure 5 shown. Taking the two-dimensional state space as an example, the state space is discretized into several (16 in this embodiment) square regions (also called square areas), and each square region represents a cell. Sampling is performed in each cell, and five sampling points are evenly set in cell Z0. The region where the initial state is located is set as cell Z0, and the five sampling points of cell Z0 ( Figure 5 represented by black dots in ) are substituted into the dynamic equation of the mobile robot (also called the dynamic model of the mobile robot), and one-step mapping is performed simultaneously using the GPU parallel computing technology, reaching the regions where cells Z1, Z4, and Z5 are located respectively. Then, the sampling points of cells Z1, Z4, and Z5 are substituted into the dynamic equation of the mobile robot for one-step mapping to reach the regions where the next cells are located, and so on until the target position is reached. According to the above steps, the mapping relationship of each cell in the discrete state space can be established. Among them, the cell mapping information includes not only the mapping pair relationship, but also the corresponding objective function value of one-step mapping and the optimal planning path solution of one-step mapping.

[0047] The generalized cell mapping method selects multiple sampling points in each cell, calculates a trajectory starting from each sampling point respectively, and each cell transfers to its image cell with a certain probability during mapping. After determining all the image cells of each cell, the one-step transition probability between any two cells can be obtained, and the transition relationship between cells is equivalent to a finite Markov chain. When calculating each trajectory among the sampling points in a cell, the GPU parallel computing technology can be used. Similarly, the GPU parallel computing technology can also be applied when calculating trajectories for all cells in the state space. The application of the GPU parallel computing technology greatly improves the computing efficiency.

[0048] S6. Based on the cell mapping relationship of all controllable cells in the obtained discrete state space based on the transition probability matrix, perform an optimal search according to the objective function of the mobile robot optimal path planning model to obtain the optimal result.

[0049] In this step, the Bellman optimality principle is specifically used to perform a global optimal search according to the objective function of the mobile robot optimal path planning model to obtain the global optimal result.

[0050] In the step of using the Bellman optimality principle to perform a global optimal search according to the objective function of the mobile robot optimal path planning model to obtain the global optimal result in this embodiment, an obstacle detection function (also called a dynamic obstacle detection function) is also set to detect obstacles.

[0051] S7. Perform real-time path planning for the mobile robot according to the obtained optimal result to obtain the real-time path planning result.

[0052] Based on the obtained optimal result, real-time path planning (also called online decision-making planning) is performed on the mobile robot to obtain the real-time path planning result.

[0053] Embodiment 4: In step S6 of this embodiment, the Bellman optimal principle is adopted to perform global optimal search according to the objective function of the optimal path planning model of the mobile robot. In the step of obtaining the global optimal result, first, an obstacle detection function is set to detect obstacles, and then a subdivision method is used for local optimal search to obtain the local optimal result. Both the global optimal result and the local optimal result of this embodiment are presented in the form of a table.

[0054] The flowchart of the global optimal search in this embodiment is as Figure 6 shown. The following details the process of performing global optimal search (to obtain the global optimal result) according to the objective function of the optimal path planning model of the mobile robot using the Bellman optimal principle: The first step: Discretize the state variables and action variables in the dynamic equation of the mobile robot and obtain its discretized dynamic equation. Then, mark the cells where the initial position and the termination position are located in the discrete state space, and define the trapped cells by the constraint equation (the trapped cells are not searched during the search and are regarded as obstacle areas).

[0055] The second step: Set sampling points in each cell (except the trapped cells) in the discrete state space based on the dynamic equation of the mobile robot.

[0056] The third step: Establish an optimal path planning model for the mobile robot according to the requirements by establishing an objective function (minimizing the travel time or minimizing the travel distance).

[0057] The fourth step: Use the GPU parallel computing technology to establish a generalized cell mapping relationship for the optimal path planning model of the mobile robot in the discrete state space and record the corresponding cell mapping information (the cell mapping information includes the probability transition matrix of the mapping relationship, the corresponding evaluation function value, and the one-step planning solution).

[0058] The fifth step: Based on the Bellman optimal principle, start the search forward from the cell where the termination position is located and record the cumulative optimal generalized cell mapping information (the cumulative optimal evaluation function value and the corresponding optimal planning solution) of each cell.

[0059] The sixth step: Continue to search forward and update the cumulative optimal objective function value and the corresponding optimal trajectory planning solution.

[0060] The seventh step: Judge the end search condition. When the optimal path planning table is no longer updated or all cells have been searched, the search ends.

[0061] The eighth step: Integrate to obtain the optimal path planning table of the mobile robot.

[0062] The flowchart of the local optimal search in this embodiment is as Figure 7 shown. The process of obtaining the local optimal result by using the subdivision method for local optimal search will be described in detail below: First, set the obstacle detection function. First, judge whether there is an obstacle (whether the obstacle detection function is triggered, and the obstacle detection function is also called the detection function). If the obstacle detection function is triggered, set the current cell as the initial cell. Use the method of establishing a local state space by subdivision (also called the subdivision method) to further subdivide and discretize the state space (nearby small area) where the current state is located, forming a subdivided discrete small state space. In the subdivided discrete small state space, use the GPU parallel computing technology to establish a generalized cell mapping relationship and use the Bellman optimal principle to perform local optimal search, so as to obtain the local optimal path planning table of the mobile robot (the local optimal path planning table is also called the local optimal path table). If the obstacle detection function is not triggered, directly use the Bellman optimal principle to perform global optimal search according to the objective function of the optimal path planning model of the mobile robot to obtain the global optimal result (the global optimal result is displayed in the form of a global optimal path table, and the global optimal path table is also called the global optimal path table).

[0063] Other steps in this embodiment (step S1, step S2, step S3, step S4, step S5, and step S7) refer to Embodiment 3.

[0064] Embodiment 5: Please refer to Figure 8 shown. The present invention also provides an electronic device 100 for the real-time path planning method of a mobile robot; the electronic device 100 includes a memory 101, at least one processor 102, a computer program 103 stored in the memory 101 and operable on the at least one processor 102, and at least one communication bus 104.

[0065] The memory 101 can be used to store the computer program 103. The processor 102 realizes the steps of the real-time path planning method for the mobile robot described in Embodiment 1 by running or executing the computer program stored in the memory 101 and calling the data stored in the memory 101. The memory 101 may mainly include a program storage area and a data storage area. Among them, the program storage area can store an operating system, application programs required for at least one function (such as a sound playback function, an image playback function, etc.); the data storage area can store data created according to the use of the electronic device 100 (such as audio data, etc.). In addition, the memory 101 may include non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) card, a FlashCard, at least one magnetic disk storage device, a flash memory device, or other non-volatile solid-state storage devices.

[0066] The at least one processor 102 can be a Central Processing Unit (CPU), or can also be other general-purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field-Programmable Gate Arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The processor 102 can be a microprocessor or the processor 102 can also be any conventional processor, etc. The processor 102 is the control center of the electronic device 100 and connects various parts of the entire electronic device 100 through various interfaces and lines.

[0067] The memory 101 in the electronic device 100 stores a plurality of instructions to implement the real-time path planning method for the mobile robot. The processor 102 can execute the plurality of instructions to thereby implement: Determine the observation space, state space, and action space of the mobile robot; Construct a dynamic model of the mobile robot according to the observation space, state space, and action space of the mobile robot, and set constraint conditions, where the constraint conditions include kinematic constraints of the mobile robot, boundary conditions, and obstacle constraints during the movement of the mobile robot; Construct an optimal path planning model for the mobile robot from the initial position to the target position, where the optimal path planning model for the mobile robot takes minimizing the travel time or minimizing the travel distance as the objective function; Discretize the state variables and control variables in the dynamic model of the mobile robot in the state space to obtain a discrete state space; Establish a generalized cell mapping for each state variable and control variable of the dynamic model, the constraint condition, and the objective function on the discrete state space to obtain the cell mapping relationship based on the transition probability matrix of all controllable cells in the discrete state space; Based on the obtained cell mapping relationship based on the transition probability matrix of all controllable cells in the discrete state space, perform an optimal search according to the objective function of the optimal path planning model of the mobile robot to obtain an optimal result; Perform real-time path planning on the mobile robot according to the obtained optimal result to obtain a real-time path planning result.

[0068] Embodiment 6: If the modules / units integrated in the electronic device 100 are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, to implement all or part of the processes in the above method embodiments of the present invention, it can also be completed by a computer program instructing relevant hardware. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, the steps of the above method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file, or some intermediate form. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, and read-only memory (ROM, Read-Only Memory).

[0069] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0070] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of the present invention. It should be understood that each flow and / or block in the flowchart and / or block diagram, and the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices produce means for implementing the functions specified in one flow Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0071] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory produce a manufactured article including instruction means that implement the functions specified in one flow Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0072] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, so that the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one flow Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0073] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: modifications or equivalent substitutions can still be made to the specific embodiments of the present invention. Any modification or equivalent substitution that does not depart from the spirit and scope of the present invention shall be covered by the protection scope of the claims of the present invention.

Claims

1. A real-time path planning method for a mobile robot, characterized in that, It includes the following steps: Determine the observation space, state space, and action space of the mobile robot; Construct a dynamic model of the mobile robot according to the observation space, state space, and action space of the mobile robot, and set constraint conditions, where the constraint conditions include kinematic constraints of the mobile robot, boundary conditions, and obstacle constraints during the movement of the mobile robot; Construct an optimal path planning model of the mobile robot from the initial position to the target position, where the optimal path planning model of the mobile robot takes minimizing the travel time or minimizing the travel distance as the objective function; Discretize the state variables and control variables in the dynamic model of the mobile robot in the state space to obtain a discrete state space; Establish a generalized cell mapping for each state variable and control variable of the dynamic model, the constraint conditions, and the objective function in the discrete state space to obtain the cell mapping relationship based on the transition probability matrix of all controllable cells in the discrete state space; Based on the obtained cell mapping relationship based on the transition probability matrix of all controllable cells in the discrete state space, perform an optimal search according to the objective function of the optimal path planning model of the mobile robot to obtain an optimal result; Perform real-time path planning for the mobile robot according to the obtained optimal result to obtain a real-time path planning result.

2. The real-time path planning method for a mobile robot according to claim 1, wherein, The step of discretizing the state variables and control variables in the dynamic model of the mobile robot in the state space to obtain a discrete state space specifically includes: Discretize the state variables and control variables in the dynamic model of the mobile robot in the state space into several cells to obtain a discrete state space.

3. The real-time path planning method for a mobile robot according to claim 1, wherein The establishment of the generalized cell mapping is specifically: using the GPU parallel computing technology to establish a generalized cell mapping for each state variable and control variable of the dynamic model, the constraint conditions, and the objective function in the discrete state space.

4. The real-time path planning method of the mobile robot according to claim 1, wherein, In the step of performing an optimal search according to the objective function of the optimal path planning model of the mobile robot based on the obtained cell mapping relationship based on the transition probability matrix of all controllable cells in the discrete state space to obtain an optimal result, specifically use the Bellman optimal principle to perform a global optimal search according to the objective function of the optimal path planning model of the mobile robot to obtain a global optimal result.

5. The real-time path planning method of the mobile robot according to claim 4, characterized in that In the step of using the Bellman optimal principle to perform a global optimal search according to the objective function of the optimal path planning model of the mobile robot to obtain a global optimal result, an obstacle detection function is also set to detect obstacles.

6. The real-time path planning method for a mobile robot according to claim 5, wherein In the step of using the Bellman optimal principle to perform a global optimal search according to the objective function of the optimal path planning model of the mobile robot to obtain a global optimal result, first set an obstacle detection function to detect obstacles, and then use a subdivision method to perform a local optimal search to obtain a local optimal result.

7. The real-time path planning method of the mobile robot according to claim 1, characterized in that, The optimal result is presented in the form of a table.

8. A real-time path planning system for a mobile robot, characterized in that, It includes a space determination module, a dynamic model construction module, an optimal path planning model construction module, a discretization module, a generalized cell mapping module, an objective function optimal search module, and a path planning module; The space determination module is used to determine the observation space, state space, and action space of the mobile robot; The dynamic model construction module is used to construct the dynamic model of the mobile robot according to the observation space, state space and action space of the mobile robot, and set constraint conditions, where the constraint conditions include the kinematic constraints of the mobile robot, boundary conditions and obstacle constraints during the movement of the mobile robot; The optimal path planning model construction module is used to construct the optimal path planning model of the mobile robot from the initial position to the target position, and the optimal path planning model of the mobile robot takes minimizing the travel time or minimizing the travel distance as the objective function; The discretization module is used to discretize the state variables and control variables in the dynamic model of the mobile robot in the state space to obtain a discrete state space; The generalized cell mapping module is used to establish a generalized cell mapping for each state variable and control variable of the dynamic model, the constraint conditions and the objective function in the discrete state space to obtain the cell mapping relationship based on the transition probability matrix of all controllable cells in the discrete state space; The objective function optimal search module is used to perform optimal search according to the objective function of the optimal path planning model of the mobile robot based on the cell mapping relationship based on the transition probability matrix of all controllable cells in the obtained discrete state space to obtain the optimal result; The path planning module is used to perform real-time path planning on the mobile robot according to the obtained optimal result to obtain the real-time path planning result.

9. An electronic device, comprising a memory and a processor, the memory storing a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the mobile robot real-time path planning method described in any one of claims 1 to 7.

10. A storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the mobile robot real-time path planning method described in any one of claims 1 to 7.

Images