Unmanned vehicle motion control method and system based on electronic and electrical information architecture

Abstract

The invention discloses an unmanned vehicle motion control method and system based on electronic and electrical information architecture, and relates to the technical field of automatic driving, and the method comprises the steps: obtaining the vehicle state information and vehicle parameters of an unmanned vehicle; establishing a kinematic model of the unmanned vehicle according to the vehicle state information; according to the vehicle parameters, a self-adaptive three-ring collision detection model of the unmanned vehicle is established; based on the kinematic model, the self-adaptive three-loop collision detection model and the electronic and electrical architecture model, estimating accumulated loop time delays of a plurality of nodes and links in the electronic and electrical architecture by adopting a multi-hop loop delay analysis method to obtain a boundary value of the loop delay; based on the boundary value of the loop delay, an improved double reinforcement learning algorithm is adopted, control training is conducted on the unmanned vehicle, and a motion control model of the unmanned vehicle is obtained.The problem of loop delay of an electronic and electrical architecture heterogeneous topology is solved, and control stability and efficient performance of the unmanned vehicle in the aspects of path tracking and safe obstacle avoidance are ensured.

Description

Technical Field

[0001] The present application relates to the field of autonomous driving technology, and in particular to a method and system for controlling the motion of an unmanned vehicle based on an electronic and electrical information architecture. Background Art

[0002] In the development of key technologies for autonomous driving, motion control systems play a vital role in the process of performing specific driving tasks. The target vehicle needs to generate a control strategy based on the reference path and real-time external environment information to achieve the goal of path tracking and safe obstacle avoidance. With the development of the concept of artificial intelligence, machine learning algorithms are constantly being applied to the motion control optimization of autonomous vehicles. Q-learning, as a representative algorithm of reinforcement learning, has been repeatedly applied to motion control problems due to its relatively simple logical structure. However, the Q-learning algorithm has the problem of unstable training effect due to the overestimation of action values, such as slow convergence or premature convergence, resulting in a non-negligible error between the estimated results and the actual results, which ultimately affects the optimization effect of the motion control strategy of autonomous driving unmanned vehicles. At the same time, most of the existing motion control research focuses on the path tracking problem, without fully considering that there is also a situation in which obstacles suddenly appear during the path tracking process, resulting in the path being blocked. Therefore, the coupling problem of path tracking control and dynamic collision avoidance control needs to be studied urgently.

[0003] In addition, relying on the advantages of powerful software updates, reduced wiring and functional integration, domain-centralized electrical and electronic architecture is becoming another hot spot in the development of intelligent vehicles. Autonomous vehicles using domain-centralized electrical and electronic architecture are an important branch of intelligent development. However, the increasing number of architectural functional components, various protocols, different network topologies and inter-domain communication modes have led to unpredictable cross-domain multi-hop information delays and system uncertainty problems. These factors are likely to affect system performance and undermine the overall system stability of autonomous vehicles. In particular, the motion control problem of autonomous vehicles involves high-precision path tracking and real-time dynamic obstacle avoidance. In practical applications, large heterogeneous loop delays cannot be ignored. Therefore, it is very necessary to analyze and solve the motion control stability problem caused by heterogeneous topology loop delays in domain-centralized electrical and electronic structures. Summary of the invention

[0004] The purpose of this application is to provide an unmanned vehicle motion control method and system based on an electronic and electrical information architecture, which can solve the control stability problem caused by the heterogeneous topology loop delay of the electronic and electrical architecture.

[0005] To achieve the above objectives, this application provides the following solutions:

[0006] In a first aspect, the present application provides a motion control method for an unmanned vehicle based on an electronic and electrical information architecture, comprising:

[0007] Obtain the vehicle status information and vehicle parameters of the unmanned vehicle.

[0008] A kinematic model of the unmanned vehicle is established according to the vehicle state information of the unmanned vehicle.

[0009] An adaptive three-ring collision detection model of the unmanned vehicle is established according to the vehicle parameters of the unmanned vehicle; the collision detection area in the adaptive three-ring collision detection model is obtained by establishing three circles with the center of the front axle of the unmanned vehicle, the center of the rear axle of the vehicle and the center of the vehicle as the respective centers of the circles.

[0010] Based on the kinematic model of the unmanned vehicle, the adaptive three-ring collision detection model and the electronic and electrical architecture model, a multi-hop loop delay analysis method is used to estimate the cumulative loop delay of multiple nodes and links in the electronic and electrical architecture to obtain the boundary value of the loop delay; the electronic and electrical architecture model is composed of a perception system, a positioning system, an intelligent network system, a switch, an on-board network, an automatic driving DCU, a chassis DCU and several domain controller units; the perception system is used to perceive the dynamic information of the external environment of the unmanned vehicle; the positioning system is used to receive the positioning information of the unmanned vehicle and the global scene map; the intelligent network system is used to receive information in the entire traffic network, and the automatic driving DCU is used to perform intelligent driving functions; the chassis DCU is used for operation control of the chassis of the unmanned vehicle; the domain controller unit is used to include a drive unit, a brake unit and a steering unit.

[0011] Based on the boundary value of the loop delay, an improved dual reinforcement learning algorithm is used to perform control training on the unmanned vehicle to obtain a motion control model of the unmanned vehicle.

[0012] In the second aspect, the present application provides an unmanned vehicle motion control system based on an electronic and electrical information architecture, including:

[0013] The parameter acquisition module is used to obtain the vehicle status information and vehicle parameters of the unmanned vehicle.

[0014] The kinematic model building module is used to build a kinematic model of the unmanned vehicle according to the vehicle state information of the unmanned vehicle.

[0015] The collision detection model establishment module is used to establish an adaptive three-ring collision detection model of the unmanned vehicle according to the vehicle parameters of the unmanned vehicle; the collision detection area in the adaptive three-ring collision detection model is obtained by establishing three circles with the center of the front axle of the unmanned vehicle, the center of the rear axle of the vehicle and the center of the vehicle as the respective centers of the circles.

[0016] A computing module is used to estimate the accumulated loop delays of multiple nodes and links in the electronic and electrical architecture based on the kinematic model, adaptive three-ring collision detection model and electronic and electrical architecture model of the unmanned vehicle, and adopt a multi-hop loop delay analysis method to obtain the boundary value of the loop delay; the electronic and electrical architecture model is composed of a perception system, a positioning system, an intelligent network system, a switch, an on-board network, an automatic driving DCU, a chassis DCU and several domain controller units; the perception system is used to perceive the dynamic information of the external environment of the unmanned vehicle; the positioning system is used to receive the positioning information of the unmanned vehicle and the global scene map; the intelligent network system is used to receive information in the entire traffic network, and the automatic driving DCU is used to perform intelligent driving functions; the chassis DCU is used for the operation control of the chassis of the unmanned vehicle; the domain controller unit is used to include a drive unit, a brake unit and a steering unit.

[0017] The motion control model building module is used to control the unmanned vehicle based on the boundary value of the loop delay and adopt an improved dual reinforcement learning algorithm to obtain the motion control model of the unmanned vehicle.

[0018] According to the specific embodiments provided in this application, this application discloses the following technical effects:

[0019] The present application provides a method and system for motion control of an unmanned vehicle based on an electronic and electrical information architecture, the method involving the following steps: first, obtaining the vehicle state information and vehicle parameters of the unmanned vehicle; then, using the vehicle state information to construct a kinematic model of the unmanned vehicle; then, establishing an adaptive three-ring collision detection model of the unmanned vehicle based on the vehicle parameters; based on the kinematic model, the adaptive three-ring collision detection model and the electronic and electrical architecture model, a multi-hop loop delay analysis method is used to estimate the cumulative loop delay of multiple nodes and links in the electronic and electrical architecture, thereby determining the boundary value of the loop delay; finally, based on the boundary value of the loop delay, an improved dual reinforcement learning algorithm is used to control and train the unmanned vehicle to obtain a motion control model of the unmanned vehicle. The present application effectively solves the problem of heterogeneous topology loop delay in the electronic and electrical architecture, and ensures the control stability and high efficiency of the unmanned vehicle in path tracking and safe obstacle avoidance. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0021] Figure 1A flowchart of an unmanned vehicle motion control method based on an electronic and electrical information architecture is provided in accordance with an embodiment of the present application.

[0022] Figure 2 A motion control model of an autonomous driving vehicle is provided in accordance with an embodiment of the present application.

[0023] Figure 3 An adaptive collision detection model for an autonomous driving vehicle is provided in accordance with an embodiment of the present application.

[0024] Figure 4 A domain-centralized electronic and electrical architecture model for an unmanned vehicle is provided in accordance with an embodiment of the present application.

[0025] Figure 5 A schematic diagram of multi-hop loop delay analysis of a domain-centralized electronic and electrical architecture provided in one embodiment of the present application.

[0026] Figure 6 A structural diagram of an unmanned vehicle motion control framework of a domain-centralized electronic and electrical information architecture provided in one embodiment of the present application.

[0027] Figure 7 A schematic diagram of the functional modules of an unmanned vehicle motion control system based on an electronic and electrical information architecture provided in one embodiment of the present application. DETAILED DESCRIPTION

[0028] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0029] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0030] Embodiment 1

[0031] like Figure 1 As shown, this embodiment provides a motion control method for an unmanned vehicle based on an electronic and electrical information architecture, including:

[0032] Step 101: Obtain vehicle status information and vehicle parameters of the unmanned vehicle.

[0033] Step 102: Establish a kinematic model of the unmanned vehicle based on the vehicle state information of the unmanned vehicle.

[0034] Step 103: Establish an adaptive three-ring collision detection model of the unmanned vehicle based on the vehicle parameters of the unmanned vehicle; the collision detection area in the adaptive three-ring collision detection model is obtained by establishing three circles with the center of the front axle of the unmanned vehicle, the center of the rear axle of the vehicle and the center of the vehicle as the respective centers.

[0035] Step 104: Based on the kinematic model of the unmanned vehicle, the adaptive three-ring collision detection model and the electronic and electrical architecture model, a multi-hop loop delay analysis method is used to estimate the cumulative loop delay of multiple nodes and links in the electronic and electrical architecture to obtain the boundary value of the loop delay; the electronic and electrical architecture model is composed of a perception system, a positioning system, an intelligent network system, a switch, an on-board network, an automatic driving DCU, a chassis DCU and several domain controller units; the perception system is used to perceive the dynamic information of the external environment of the unmanned vehicle; the positioning system is used to receive the positioning information of the unmanned vehicle and the global scene map; the intelligent network system is used to receive information in the entire traffic network, and the automatic driving DCU is used to perform intelligent driving functions; the chassis DCU is used for operation control of the chassis of the unmanned vehicle; the domain controller unit is used to include a drive unit, a brake unit and a steering unit.

[0036] Step 105: Based on the boundary value of the loop delay, an improved dual reinforcement learning algorithm is used to perform control training on the unmanned vehicle to obtain a motion control model of the unmanned vehicle.

[0037] When executing step 102, the specific steps may be as follows:

[0038] According to the vehicle state information (coordinate information) of the unmanned vehicle, based on the front and rear axle constraints and geometric equations, the kinematic model of the unmanned vehicle motion control can be expressed as the following state space equation:

[0039]

[0040] like Figure 2 As shown in the figure, XOY and xoy represent the inertial coordinate system and the local coordinate system respectively. In the inertial coordinate system, the center coordinates of the front axle and the rear axle of the vehicle are represented by (X f ,Y f ) and (X r ,Y r ) represents the yaw angle of the vehicle. In the local coordinate system, v r and δ f They represent the speed of the rear axle center and the steering angle of the front axle respectively. In addition, l is the vehicle wheelbase and R is the instantaneous turning radius of the rear axle center.

[0041] When executing step 103, the specific steps may be as follows:

[0042] During the motion control process of autonomous driving vehicles, real-time collision detection is required to ensure driving safety.

[0043] This embodiment constructs an adaptive three-ring collision detection model, such as Figure 3 Assume that the speed of the target vehicle is 0m / s, and the collision detection area is Figure 3 The original area shown on the right. Based on the geometric center of the vehicle, three rings are arranged into an array to cover the actual physical area occupied by the vehicle. Once an obstacle enters the collision detection area, it is determined that a collision must occur.

[0044] Assuming that the target vehicle is traveling at a certain speed, the motion state information of the unmanned vehicle needs to be considered to dynamically adjust the boundary of the collision detection area and determine the real-time size of the detection area by calculating the radius of the adaptive circle. The calculation formula is as follows:

[0045]

[0046] Where r is the original radius of the collision detection area, and κ is the weight factor of the adjustable length. ini and a brk,max Represent the initial driving speed and maximum braking acceleration respectively. Braking time t brk By v ini and a brk,max If there is no other object in the adaptive collision detection area, the driving state is determined to be a safe state.

[0047] Before executing step 104, the method further includes:

[0048] Modeling of domain-centralized electrical and electronic architecture on unmanned vehicles, including:

[0049] The domain-centralized electronic and electrical architecture used in autonomous driving vehicles is a typical network-physical system, which realizes the connection and interaction between the network side and the physical side.

[0050] The cyber-physical system includes a perception system, a positioning system, intelligent network components, switches, vehicle networks, and multiple domain controller units (DCUs). Among them, the perception system, positioning system, and intelligent network components are connected to the autonomous driving DCU via Ethernet. Multiple domain controller units represented by the autonomous driving DCU are connected to the switch via Ethernet. The drive unit, brake unit, and steering unit are connected to the chassis DCU via CAN lines.

[0051] Among the components involved in this embodiment, the perception system is responsible for perceiving scene information and the vehicle's own status information; the positioning system is responsible for receiving the vehicle's positioning information; the intelligent network connection element is responsible for receiving the driving status information of other vehicles in the external environment; the autonomous driving DCU is responsible for realizing various intelligent driving functions of the autonomous driving vehicle, including motion control functions; the switch is responsible for realizing information transmission between different DCUs; the chassis DCU is responsible for realizing the operation control of the chassis of the autonomous driving vehicle, and can convert the received control commands into signals that can be recognized by the actuator unit; the drive unit is responsible for the acceleration drive of the vehicle; the braking unit is responsible for the deceleration braking of the vehicle; and the steering unit is responsible for the driving and steering of the vehicle.

[0052] The electronic and electrical architecture in this embodiment includes multiple service nodes and links, such as Figure 4 As shown in Figure 1. In terms of network, the architecture includes perception systems (sensor units), positioning systems (positioning units), intelligent connected components (intelligent connected units), switches, vehicle networks, and multiple domain controller units (DCUs). In terms of physics, the architecture includes vehicle mechanical components such as chassis actuators, etc.

[0053] During the operation of the electronic and electrical architecture, the specific DCU is responsible for the main functions of the corresponding functional area. In particular, the motion controller proposed in this embodiment is integrated into the autonomous driving DCU to collaboratively achieve the tasks of path tracking and safe obstacle avoidance. The switch serves as a data exchange center to complete the interaction between different domain controllers.

[0054] In the process of realizing complete autonomous driving unmanned vehicle motion control, the perception system receives external environment information, vehicle status and reference path, and sends it to the autonomous driving DCU via Ethernet. Then, the motion control system proposed in this embodiment is embedded in the autonomous driving DCU, and the optimal motion control command is solved, and then sent to the chassis DCU using the transmission capacity of the switch. Then, the chassis DCU converts the control command into a signal that can be recognized by the actuator unit. Finally, the actuator unit applies drive, braking and steering control to the autonomous driving vehicle according to the control signal. Figure 5 As shown in the figure, the whole process consists of 5 service nodes (perception unit (sensor unit), autonomous driving DCU, switch, chassis DCU, actuator unit) and 4 communication links, and cross-domain information transmission is completed through 4 jumps (transmission #1 to transmission #4).

[0055] When executing step 104, the specific steps may be as follows:

[0056] First, the service time of nodes and links is calculated, including:

[0057] In the electronic and electrical architecture, the control loop consists of 5 service nodes and 4 communication links, and cross-domain information transmission requires 4 jumps. Therefore, loop delays in heterogeneous topologies are inevitable and will have a non-negligible negative impact on motion control performance. To address this problem, this embodiment uses a multi-hop loop delay analysis method to estimate the cumulative loop delays of multiple nodes and links in a domain-centralized electronic and electrical architecture.

[0058] The accumulation of loop delay is directly related to the service time of each network component (network link or network node). Service time is the time required for a node or link to complete the relevant function, and the calculation method is:

[0059]

[0060] Among them, the node τ node The service time includes the task queuing time τ tq and task execution time τ ti , link τ link The service time includes the message queuing time τ mq and the message transmission time τ mt .

[0061] Then, the cumulative delay is calculated as follows:

[0062] For the complete control loop under the electronic and electrical architecture of the autonomous driving unmanned vehicle, the cumulative loop delay τ can be derived from the service time of 5 service nodes and 4 communication links. delay :

[0063]

[0064] Where m is equal to 5, n is equal to 4, τ node,i is the i-th node, τ link,j is the jth link.

[0065] Finally, according to the driving mode of these service nodes (i.e., time-driven mode TD or event-driven mode ED) and the triggering mode of these communication links (time-triggered mode TT or event-triggered mode ET), the theoretical bound of the loop delay at any time k can be expressed as:

[0066]

[0067] In the formula, τ node,i,k is the delay of the ith node at time k, τ link,j,k is the delay of the jth link at time k, τ tq,i,k represents the task queuing time of the i-th node at time k, τ ti,i,k represents the task execution time of the i-th node at time k, τmq,j,k represents the message queuing time at time k in the jth link, τ mt,j,k Indicates the message transmission time at time k in the jth link. TD mode is time-driven mode, ED mode is event-driven mode, TT mode is time-triggered mode, and ETmode is event-triggered mode.

[0068] Based on the above formula, further, the multi-hop loop delay analysis diagram of the motion control process using the domain-centralized electronic and electrical architecture is as follows: Figure 5 shown.

[0069] Therefore, the theoretical limit of the loop delay at time k can be obtained:

[0070]

[0071] Among them, T s is the sampling time, τ node,i is the delay of the ith node, τ link,j is the delay of the jth link, m is the total number of nodes, n is the total number of links, τ tq is the task queuing time, τ ti is the task execution time, τ mq is the message queuing time, τ mt is the message transmission time, k is the time, τ tq,1,k is the task queuing time of the sensing system at time k, τ ti,1,k is the task execution time of the perception system at time k, T s is the sampling time, SUP represents the supremum of the summation, τ mq,4,k is the message queuing time from the chassis DCU to the actuator unit in the domain controller unit, τ mt,4,k is the message transmission time from chassis DCU to actuator unit, τ tq,5,k is the task queuing time of the executor unit at time k, τ ti,5,k is the task execution time of the actuator unit at time k. In the training optimization of the motion control system, the estimated electronic and electrical architecture loop delay value will be integrated into the algorithm design of the reinforcement learning system.

[0072] Before executing step 105, the method further includes:

[0073] Design an improved double reinforcement (double Q) learning algorithm, specifically including:

[0074] Get two Q functions.

[0075] The two Q functions are trained by cross-learning method to obtain an improved dual reinforcement learning algorithm.

[0076] Specifically, this embodiment designs an improved dual Q learning algorithm, which uses two Q functions to evaluate the value of the motion control strategy, and adopts a cross-learning approach to train and update the two Q functions to solve the problem of overestimation of action values.

[0077] Specifically, when executing step 105, it can be as follows:

[0078] like Figure 6 As shown, an improved dual reinforcement learning algorithm is used to reconstruct the reward function of reinforcement learning based on the boundary value of the loop delay; the reconstructed reward function is optimized and iterated to obtain the motion control model of the unmanned vehicle.

[0079] Specifically, based on the improved dual reinforcement (dual Q) learning algorithm, the estimated loop delay number boundary value is integrated into the reconstruction of the reinforcement learning reward function to solve the control stability problem caused by heterogeneous topology loop delay in the domain centralized electronic and electrical architecture. Through continuous optimization training, a stable and efficient motion control system can be finally obtained.

[0080] Among them, in the traditional reinforcement learning algorithm, the reward function can be expressed as:

[0081]

[0082] Among them, k acu , k saf , k rap and k com They are accuracy coefficient, safety coefficient, rapidity coefficient and comfort coefficient respectively. tra 、f obs 、f vel 、f dri They represent the tracking error function, safety obstacle avoidance function, speed evaluation function and comfortable driving function respectively.

[0083] In order to eliminate the interference of the electronic and electrical architecture loop delay during the execution of the control strategy, the time difference between the state and the control is determined according to the theoretical boundary value of the loop delay, and then the reward function is reconstructed:

[0084]

[0085] Among them, r ori represents the initial reward function, f τdelay represents the execution function considering the delay, u t-τdelay represents the control action considering the delay, P reference represents the reference path, Obcs represents the obstacle information, R adaptive Represents the adaptive radius of vehicle collision detection, is the accumulated loop delay.

[0086] Furthermore, the objective value function and optimal value function of the motion control problem can be expressed as:

[0087]

[0088] In the formula, ξ t represents the weight set of the Q matrix. The update mechanism of the Q matrix is ​​as follows:

[0089]

[0090] Where α and γ represent the learning rate and discount factor respectively. The training of the Q matrix is ​​to update the weight set ξ t The completeness of the weight set directly means the completeness of the Q matrix. This embodiment introduces the concept of double estimation and uses two Q matrices (Q A ,Q B ) is used to evaluate the value of the control strategy. The optimal control action to be performed is obtained by maximizing the mean of the two Q functions. The implementation formula is as follows:

[0091]

[0092] In the formula, ξ t A ,ξ t B is the weight set of the Q matrix.

[0093] The update of the two Q matrices uses different sample experience sets, and the internal weight update of the Q matrix is ​​realized by cross-learning. The update process can be expressed as:

[0094]

[0095] Then, in order to improve the training efficiency of the Q matrix in the reinforcement learning system, priority experience replay (PER) is used to select experience samples for training. Compared with traditional random experience replay, PER training is more targeted, and the sampling probability of each experience sample can be calculated by the following formula:

[0096]

[0097] Where TD(s,u) is the temporal difference error (TD), which is used to evaluate the error between the target value function and the actual value function of a specific experience sample. The selection priority of the sample is positively correlated with the TD error value. σ is used to avoid division failure, and μ represents the coefficient power of the priority. When the value function error approaches 0 infinitely, it means that the training of the reinforcement learning system proposed in this embodiment is approaching perfection.

[0098] Finally, if Figure 6 As shown, the reference path, vehicle status and external environment information are input into the trained motion control system together, and the vehicle's driving acceleration and steering angle are output to the execution system as control commands through the electronic and electrical architecture.

[0099] Embodiment 2

[0100] like Figure 7 As shown, this embodiment provides an unmanned vehicle motion control system based on an electronic and electrical information architecture, including:

[0101] The parameter acquisition module 701 is used to obtain the vehicle status information and vehicle parameters of the unmanned vehicle.

[0102] The kinematic model building module 702 is used to build a kinematic model of the unmanned vehicle according to the vehicle state information of the unmanned vehicle.

[0103] The collision detection model establishment module 703 is used to establish an adaptive three-ring collision detection model of the unmanned vehicle according to the vehicle parameters of the unmanned vehicle; the collision detection area in the adaptive three-ring collision detection model is obtained by establishing three circles with the center of the front axle of the unmanned vehicle, the center of the rear axle of the vehicle and the center of the vehicle as the respective centers of the circles.

[0104] The computing module 704 is used to estimate the accumulated loop delays of multiple nodes and links in the electronic and electrical architecture based on the kinematic model, adaptive three-ring collision detection model and electronic and electrical architecture model of the unmanned vehicle, and obtain the boundary value of the loop delay by adopting a multi-hop loop delay analysis method; the electronic and electrical architecture model is composed of a perception system, a positioning system, an intelligent network system, a switch, an on-board network, an automatic driving DCU, a chassis DCU and several domain controller units; the perception system is used to perceive the dynamic information of the external environment of the unmanned vehicle; the positioning system is used to receive the positioning information of the unmanned vehicle and the global scene map; the intelligent network system is used to receive information in the entire traffic network, and the automatic driving DCU is used to perform intelligent driving functions; the chassis DCU is used for the operation control of the chassis of the unmanned vehicle; the domain controller unit is used to include a drive unit, a brake unit and a steering unit.

[0105] The motion control model building module 705 is used to perform control training on the unmanned vehicle based on the boundary value of the loop delay and adopt an improved dual reinforcement learning algorithm to obtain the motion control model of the unmanned vehicle.

[0106] The motion control model building module 705 specifically includes:

[0107] The reconstruction submodule is used to reconstruct the reward function of reinforcement learning based on the boundary value of the loop delay by adopting an improved dual reinforcement learning algorithm.

[0108] The model building submodule is used to optimize and iterate the reconstructed reward function to obtain the motion control model of the unmanned vehicle.

[0109] In summary, this application has the following technical effects:

[0110] 1) Aiming at the problem of efficient motion control of unmanned vehicles with domain-centralized electronic and electrical architecture, the method provided in this application significantly solves the control stability problem caused by the heterogeneous topology loop delay of the electronic and electrical architecture, while ensuring the high efficiency in path tracking and safe obstacle avoidance performance of autonomous driving unmanned vehicles.

[0111] 2) This application innovatively introduces a multi-hop loop delay analysis method, which accurately estimates the theoretical boundary value of the closed-loop time lag and deeply integrates it into the controller optimization process, thereby effectively responding to and solving the challenges of control stability.

[0112] 3) This application applies the optimized dual Q learning algorithm to the training process of the motion control system, which effectively avoids the problem of overestimation of action values ​​that may be encountered in the optimization stage of the traditional Q learning method, thereby achieving a significant increase in the training optimization speed and a significant enhancement in the training effect.

[0113] 4) The unmanned vehicle motion control method based on the domain-centralized electronic and electrical architecture proposed in this application realizes the deep integration of the information communication mechanism of the electronic and electrical architecture with traditional motion control problems, and provides solid support for fully autonomous motion control in a driverless environment. This method has demonstrated excellent performance in multiple dimensions such as tracking accuracy, obstacle avoidance safety, driving speed and comfort.

[0114] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0115] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. At the same time, for those skilled in the art, according to the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A motion control method for an unmanned vehicle based on an electronic and electrical information architecture, characterized in that: The unmanned vehicle motion control method based on the electronic and electrical information architecture includes: Obtain vehicle status information and vehicle parameters of the unmanned vehicle; Establishing a kinematic model of the unmanned vehicle according to the vehicle state information of the unmanned vehicle; According to the vehicle parameters of the unmanned vehicle, an adaptive three-ring collision detection model of the unmanned vehicle is established; the collision detection area in the adaptive three-ring collision detection model is obtained by establishing three circles with the center of the front axle of the unmanned vehicle, the center of the rear axle of the vehicle and the center of the vehicle as the respective centers; Based on the kinematic model, adaptive three-ring collision detection model and electronic and electrical architecture model of the unmanned vehicle, a multi-hop loop delay analysis method is used to estimate the cumulative loop delay of multiple nodes and links in the electronic and electrical architecture to obtain the boundary value of the loop delay; the electronic and electrical architecture model is composed of a perception system, a positioning system, an intelligent network system, a switch, an on-board network, an automatic driving DCU, a chassis DCU and several domain controller units; the perception system is used to perceive the dynamic information of the external environment of the unmanned vehicle; the positioning system is used to receive the positioning information of the unmanned vehicle and the global scene map; the intelligent network system is used to receive information in the entire traffic network, and the automatic driving DCU is used to perform intelligent driving functions; the chassis DCU is used for the operation control of the chassis of the unmanned vehicle; the domain controller unit is used to include a drive unit, a brake unit and a steering unit; Based on the boundary value of the loop delay, an improved dual reinforcement learning algorithm is used to perform control training on the unmanned vehicle to obtain a motion control model of the unmanned vehicle.

2. The unmanned vehicle motion control method based on electronic and electrical information architecture according to claim 1 is characterized in that: The formula expression of the kinematic model is specifically: Among them, in the inertial coordinate system, (X f ,Y f ) is the center coordinate of the front axle of the vehicle, (X r ,Y r ) represents the center coordinate of the rear axle, represents the yaw angle of the vehicle; in the local coordinate system, v r and δ f They represent the speed of the rear axle center and the steering angle of the front axle respectively, l is the vehicle wheelbase, R is the instantaneous turning radius of the rear axle center, and ω represents the steering angular velocity of the vehicle.

3. The unmanned vehicle motion control method based on electronic and electrical information architecture according to claim 1 is characterized in that: The calculation formula of the adaptive circle radius in the adaptive three-ring collision detection model is: Where r is the original radius of the collision detection area, κ is the weight factor of the adjustable length, and v ini and a brk,max represent the initial driving speed and the maximum braking acceleration respectively, t brk is the braking time.

4. The unmanned vehicle motion control method based on electronic and electrical information architecture according to claim 1 is characterized in that: Based on the kinematic model of the unmanned vehicle, the adaptive three-ring collision detection model and the electronic and electrical architecture model, a multi-hop loop delay analysis method is used to estimate the cumulative loop delay of multiple nodes and links in the electronic and electrical architecture, and obtain the boundary value of the loop delay, which specifically includes: According to the formula Calculate the cumulative round-trip delay of multiple nodes and links; Based on the accumulated loop delay, according to the formula Calculate the boundary value of the loop delay; Among them, T s is the sampling time, τ node,i is the delay of the ith node, τ link,j is the delay of the jth link, m is the total number of nodes, n is the total number of links, τ tq is the task queuing time, τ ti is the task execution time, τ mq is the message queuing time, τ mt is the message transmission time, k is the time, τ tq,1,k is the task queuing time of the sensing system at time k, τ ti,1,k is the task execution time of the perception system at time k, T s is the sampling time, SUP represents the supremum of the summation, τ mq,4,k is the message queuing time from the chassis DCU to the actuator unit in the domain controller unit, τ mt,4,k is the message transmission time from chassis DCU to actuator unit, τ tq,5,k is the task queuing time of the executor unit at time k, τ ti,5,k is the task execution time of the executor unit at time k.

5. The unmanned vehicle motion control method based on electronic and electrical information architecture according to claim 1 is characterized in that: The improvement steps of the improved dual reinforcement learning algorithm are: Get two Q functions; The two Q functions are trained by cross-learning method to obtain an improved dual reinforcement learning algorithm.

6. The unmanned vehicle motion control method based on electronic and electrical information architecture according to claim 1 is characterized in that: Based on the boundary value of the loop delay, an improved dual reinforcement learning algorithm is used to perform control training on the unmanned vehicle to obtain a motion control model of the unmanned vehicle, which specifically includes: An improved dual reinforcement learning algorithm is used to reconstruct the reward function of reinforcement learning based on the boundary value of the loop delay; The reconstructed reward function is optimized and iterated to obtain the motion control model of the unmanned vehicle.

7. The unmanned vehicle motion control method based on electronic and electrical information architecture according to claim 6 is characterized in that: The expressions of the objective value function and the optimal value function of the improved dual reinforcement learning algorithm are: Among them, ξ t represents the weight set of the Q matrix, Indicates the boundary value of the loop delay, s t Indicates the current state, u t represents the control action at the current moment, and γ represents the discount factor of the reinforcement learning algorithm.

8. The unmanned vehicle motion control method based on electronic and electrical information architecture according to claim 7 is characterized in that: The formula expression of the reconstructed reward function is: Among them, r ori represents the initial reward function, f τdelay represents the execution function considering the delay, u t-τdelay represents the control action considering the delay, P reference represents the reference path, Obcs represents the obstacle information, R adaptive Represents the adaptive radius of vehicle collision detection, is the accumulated loop delay.

9. An unmanned vehicle motion control system based on electronic and electrical information architecture, characterized in that: include: A parameter acquisition module is used to obtain vehicle status information and vehicle parameters of the unmanned vehicle; A kinematic model building module, used to build a kinematic model of the unmanned vehicle according to the vehicle state information of the unmanned vehicle; A collision detection model establishment module is used to establish an adaptive three-ring collision detection model of the unmanned vehicle according to the vehicle parameters of the unmanned vehicle; the collision detection area in the adaptive three-ring collision detection model is obtained by establishing three circles with the center of the front axle of the unmanned vehicle, the center of the rear axle of the vehicle and the center of the vehicle as the respective centers; A computing module is used to estimate the accumulated loop delays of multiple nodes and links in the electronic and electrical architecture based on the kinematic model, adaptive three-ring collision detection model and electronic and electrical architecture model of the unmanned vehicle, and obtain the boundary value of the loop delay by adopting a multi-hop loop delay analysis method; the electronic and electrical architecture model is composed of a perception system, a positioning system, an intelligent network system, a switch, an on-board network, an automatic driving DCU, a chassis DCU and several domain controller units; the perception system is used to perceive the dynamic information of the external environment of the unmanned vehicle; the positioning system is used to receive the positioning information of the unmanned vehicle and the global scene map; The intelligent network system is used to receive information from the entire traffic network, the autonomous driving DCU is used to perform intelligent driving functions; the chassis DCU is used to control the operation of the unmanned vehicle chassis; the domain controller unit is used to include a drive unit, a brake unit and a steering unit; The motion control model building module is used to control the unmanned vehicle based on the boundary value of the loop delay and adopt an improved dual reinforcement learning algorithm to obtain the motion control model of the unmanned vehicle.

10. The unmanned vehicle motion control system based on electronic and electrical information architecture according to claim 9, characterized in that: The motion control model building module specifically includes: A reconstruction submodule, used to reconstruct the reward function of reinforcement learning based on the boundary value of the loop delay by using an improved dual reinforcement learning algorithm; The model building submodule is used to optimize and iterate the reconstructed reward function to obtain the motion control model of the unmanned vehicle.

Images