Wheel control method and apparatus for an unmanned vehicle

Abstract

The application discloses a wheel control method and device of an unmanned vehicle, and belongs to the technical field of vehicle control. The method comprises the following steps: determining target state data of the unmanned vehicle at a current time and a target driving path from a current position to a destination according to multi-source sensing data and destination position coordinates; determining the whole vehicle longitudinal driving force of the unmanned vehicle at the current time based on a multi-loop PID controller of vehicle longitudinal motion adjustment according to the target state data and corresponding current expected position coordinates of the unmanned vehicle on the target driving path; determining the whole vehicle expected yawing moment of the unmanned vehicle at the current time based on an MPC prediction model according to the target state data, the current yawing angular velocity in the multi-source sensing data, the current expected position coordinates and the current expected heading angle corresponding to the current expected position coordinates; and controlling the wheels of the unmanned vehicle according to the whole vehicle longitudinal driving force and the whole vehicle expected yawing moment.

Description

Technical Field

[0001] This invention relates to the field of vehicle control technology, and in particular to a wheel control method and device for an unmanned vehicle. Background Technology

[0002] With the development of autonomous driving technology, higher requirements are being placed on the wheel control of autonomous vehicles. Current six-wheel all-wheel drive control methods have the following problems: difficulty in wheel coordination, which easily leads to wheel conflicts; poor sideslip suppression during differential steering, resulting in insufficient vehicle driving stability; and complex multi-wheel cooperative algorithms are difficult to adapt to the high real-time requirements of onboard control. Summary of the Invention

[0003] This invention provides a wheel control method and device for autonomous vehicles to improve the multi-wheel coordination effect of autonomous vehicles, avoid the risk of exceeding speed, torque and attitude limits, and improve the driving stability and steering safety of autonomous vehicles.

[0004] According to one aspect of the present invention, a wheel control method for an unmanned vehicle is provided, the unmanned vehicle being configured with four drive wheels and two support wheels, the method comprising: Acquire multi-source perception data of the autonomous vehicle at the current moment and the destination coordinates of the autonomous vehicle; the multi-source perception data includes the current pose inertial data; Based on multi-source perception data and destination location coordinates, determine the target state data of the autonomous vehicle at the current moment, as well as the target driving path from the current location to the destination. The system obtains the current desired position coordinates of the autonomous vehicle on the target driving path, and determines the longitudinal driving force of the autonomous vehicle at the current moment based on the current desired position coordinates and target state data, using a multi-loop PID controller for longitudinal motion adjustment. The multi-loop PID controller includes a position loop PID controller, a velocity loop PID controller, and an acceleration loop PID controller. Based on the target state data, the current yaw rate, the current desired position coordinates, and the current desired heading angle corresponding to the current desired position coordinates in the current pose inertial data, the desired yaw moment of the unmanned vehicle at the current moment is determined based on the MPC prediction model. Based on the longitudinal driving force of the whole vehicle and the desired yaw moment of the whole vehicle, determine the motor torque corresponding to each drive wheel of the autonomous vehicle; The wheels of the autonomous vehicle are controlled based on the motor torque corresponding to each drive wheel.

[0005] According to another aspect of the present invention, a wheel control device for an unmanned vehicle is provided, the unmanned vehicle being configured with four drive wheels and two support wheels, the device comprising: The data acquisition module is used to acquire the multi-source perception data of the autonomous vehicle at the current moment and the destination coordinates of the autonomous vehicle; the multi-source perception data includes the current pose inertial data; The target driving path determination module is used to determine the target state data of the autonomous vehicle at the current moment, as well as the target driving path from the current location to the destination, based on multi-source perception data and destination location coordinates. The vehicle longitudinal driving force determination module is used to obtain the current expected position coordinates of the autonomous vehicle on the target driving path, and determine the vehicle longitudinal driving force at the current moment based on the current expected position coordinates and target state data, using a multi-loop PID controller for vehicle longitudinal motion adjustment; the multi-loop PID controller includes a position loop PID controller, a speed loop PID controller and an acceleration loop PID controller. The whole vehicle expected yaw moment determination module is used to determine the whole vehicle expected yaw moment of the autonomous vehicle at the current moment based on the target state data, the current yaw angular velocity in the current pose inertial data, the current expected position coordinates, and the expected heading angle corresponding to the current expected position coordinates, based on the MPC prediction model. The wheel motor torque determination module is used to determine the motor torque corresponding to each drive wheel of the autonomous vehicle based on the longitudinal driving force of the vehicle and the desired yaw torque of the vehicle. The wheel control module is used to control the wheels of the autonomous vehicle based on the motor torque corresponding to each drive wheel.

[0006] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising: At least one processor; and a memory communicatively connected to at least one processor; wherein, The memory stores a computer program that can be executed by at least one processor, such that the at least one processor is able to execute the wheel control method of an unmanned vehicle according to any embodiment of the present invention.

[0007] According to another aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions for causing a processor to execute and implement the wheel control method of an unmanned vehicle according to any embodiment of the present invention.

[0008] According to another aspect of the present invention, a computer program product is provided, comprising a computer program that, when executed by a processor, implements the wheel control method of an unmanned vehicle according to any embodiment of the present invention.

[0009] The technical solution of this invention involves acquiring multi-source perception data of an autonomous vehicle at the current moment and the destination coordinates of the autonomous vehicle; the multi-source perception data includes current pose inertial data; based on the multi-source perception data and the destination coordinates, the target state data of the autonomous vehicle at the current moment and the target driving path from the current position to the destination are determined; the current desired position coordinates of the autonomous vehicle on the target driving path are acquired, and based on the current desired position coordinates and the target state data, a multi-loop PID controller for adjusting the vehicle's longitudinal motion is used to determine the longitudinal driving force of the autonomous vehicle at the current moment; the multi-loop PID controller includes a position loop PID controller, a velocity loop PID controller, and an acceleration loop PID controller; based on the target state data, the current yaw rate in the current pose inertial data, the current desired position coordinates, and the current desired heading angle corresponding to the current desired position coordinates, the desired yaw torque of the autonomous vehicle at the current moment is determined based on an MPC prediction model; based on the longitudinal driving force and the desired yaw torque, the motor torque corresponding to each drive wheel of the autonomous vehicle is determined; and based on the motor torque corresponding to each drive wheel of the autonomous vehicle, the wheels of the autonomous vehicle are controlled. The aforementioned technical solution, by fusing multi-source perception data of the autonomous vehicle at the current moment, obtains the target state data of the autonomous vehicle at the current moment, improving the accuracy and reliability of the target state data. Simultaneously, based on the multi-source perception data of the autonomous vehicle at the current moment, a target driving path is generated, improving the accuracy of the target driving path. Then, based on the target state data and the expected vehicle state data corresponding to each planned trajectory point on the target driving path, combined with a multi-loop PID controller and MPC prediction model for longitudinal motion adjustment, wheel-to-wheel coordinated control of the autonomous vehicle is performed. The multi-loop PID controller quickly corrects operational deviations, eliminates steady-state deviations, and suppresses control oscillations, effectively resolving the problem of inter-wheel drive force interference and avoiding inter-wheel conflicts, thereby improving the multi-wheel coordination effect of the autonomous vehicle and mitigating the risks of exceeding speed, torque, and attitude limits. Furthermore, by using the MPC prediction model to predict the vehicle's driving state in advance, the expected yaw torque of the autonomous vehicle at the current moment is optimized under multiple constraints, effectively suppressing steering sideslip, thereby improving the sideslip suppression effect during differential steering of the autonomous vehicle and enhancing its driving stability and steering safety. The entire wheel control logic is streamlined and efficient, and can be well adapted to the high real-time vehicle control requirements, ensuring real-time control response of the wheels.

[0010] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0011] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 This is a flowchart of a wheel control method for an unmanned vehicle according to Embodiment 1 of the present invention; Figure 2 This is a schematic diagram of a chassis wheel assembly for an unmanned vehicle provided in an embodiment of the present invention; Figure 3 This is a flowchart of a wheel control method for an unmanned vehicle according to Embodiment 2 of the present invention; Figure 4 This is a schematic diagram of the structure of a wheel control device for an unmanned vehicle according to Embodiment 3 of the present invention; Figure 5 This is a schematic diagram of the structure of an electronic device that implements the wheel control method for an unmanned vehicle according to an embodiment of the present invention. Detailed Implementation

[0013] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0014] It should be noted that the terms "target," "candidate," "first," and "second," etc., used in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0015] Example 1 Figure 1 This is a flowchart of a wheel control method for an autonomous vehicle according to Embodiment 1 of the present invention. This embodiment is applicable to situations where the wheels of an autonomous vehicle need to be controlled. The method can be executed by a wheel control device for the autonomous vehicle, which can be implemented in hardware and / or software and can be configured in an electronic device. It should be noted that... (See also...) Figure 2 Autonomous vehicles are equipped with four drive wheels (i.e., Figure 2 Wheels No. 1, 3, 4 and 6) and two support wheels (i.e. Figure 2 The drive wheels (wheels 2 and 5) of the autonomous vehicle provide power control, while the support wheels provide structural support and load distribution, playing a lateral stabilizing role during steering and effectively suppressing sideslip. This design avoids the complexity of wheel control and high energy consumption associated with traditional autonomous vehicles where all six wheels are drive wheels, while effectively suppressing sideslip during steering. It should also be noted that each of the six wheels of the autonomous vehicle is equipped with a completely independent suspension system; each wheel has its own independent damping module consisting of shock-absorbing springs and hydraulic dampers. This design isolates rigid interference between the left and right wheels, effectively protecting the vehicle's precision sensors (such as lidar sensors) from high-frequency vibration damage and ensuring that the two support wheels and four drive wheels maintain contact with the ground under various road conditions (such as uneven surfaces), giving the autonomous vehicle excellent ground contact performance.

[0016] like Figure 1 As shown, the method includes: S101. Obtain the multi-source perception data of the autonomous vehicle at the current moment and the destination coordinates of the autonomous vehicle; the multi-source perception data includes the current pose inertial data.

[0017] Multi-source perception data refers to data collected by various onboard sensors equipped on the autonomous vehicle. Current pose inertial data refers to data collected at the current moment by the IMU (Inertial Measurement Unit) equipped on the autonomous vehicle at a preset sampling frequency (e.g., 100Hz); optionally, current pose inertial data includes current three-axis acceleration, current three-axis angular velocity, and current three-axis attitude angle. Current three-axis acceleration refers to the linear acceleration of the autonomous vehicle along the three orthogonal axes of the vehicle body at the current moment; optionally, current three-axis acceleration includes current x-axis acceleration, current y-axis acceleration, and current z-axis acceleration. Current three-axis angular velocity refers to the rotational angular velocity of the autonomous vehicle about the three orthogonal axes at the current moment; optionally, current three-axis angular velocity includes current roll angular velocity, current pitch angular velocity, and current yaw angular velocity. Current three-axis attitude angle refers to the spatial attitude angle of the autonomous vehicle body at the current moment; optionally, current three-axis attitude angle includes current roll angle, current pitch angle, and current actual heading angle.

[0018] Optionally, the multi-source perception data also includes current radar point cloud data, current satellite positioning data, and current wheel speed information. Specifically, current radar point cloud data refers to data collected at the current moment by the lidar sensor equipped on the autonomous vehicle; current satellite positioning data refers to data collected at the current moment by the Global Navigation Satellite System (GNSS) equipped on the autonomous vehicle; and current wheel speed information refers to data collected at the current moment by the wheel speed encoder equipped on the autonomous vehicle.

[0019] S102. Based on multi-source perception data and destination location coordinates, determine the target state data of the autonomous vehicle at the current moment, as well as the target driving path from the current location to the destination.

[0020] The target state data is the vehicle state data obtained by fusing multi-source perception data. The current location refers to the actual location of the autonomous vehicle at the current moment. The target travel path refers to the travel path taken by the autonomous vehicle from its current location to its destination. It should be noted that the target travel path is essentially a timestamped parameterized state sequence, represented as follows: .in, This indicates that at time t, the autonomous vehicle is expected to reach the following state: located at position (x, y) with a heading angle of θ. The linear velocity is v, and the angular velocity is... .

[0021] Specifically, data fusion processing can be performed on the current radar point cloud data and the current pose inertial data to obtain the 6DoF pose data and static point cloud data of the autonomous vehicle at the current moment; data fusion processing can be performed on the 6DoF pose data, the current pose inertial data, the current satellite positioning data, and the current wheel speed information to obtain the target state data of the autonomous vehicle at the current moment; based on the 6DoF pose data and static point cloud data, the current grid map corresponding to the autonomous vehicle at the current moment can be determined using a ray casting algorithm; the historical grid map corresponding to the autonomous vehicle at the previous moment can be obtained, and based on the historical grid map and the current grid map, the current ESDF map can be generated using a signed distance field generation algorithm; based on the target state data, the destination location coordinates, and the current ESDF map, the target driving path of the autonomous vehicle from the current position to the destination can be determined using the A* algorithm; the weights of the heuristic function in the A* algorithm are determined based on the signed Euclidean distance value corresponding to the current position point in the current ESDF map, the remaining distance from the current position point to the destination, and the local obstacle density of the current position point.

[0022] Among them, 6DoF pose data refers to six-degree-of-freedom pose data; optionally, 6DoF pose data includes three-axis position and three-axis attitude angles. Static point cloud data refers to point cloud data generated by scanning stationary objects in the environment; current grid map refers to the grid map corresponding to the autonomous vehicle at the current moment; previous moment refers to the moment before the current moment; historical grid map refers to the grid map corresponding to the autonomous vehicle at the previous moment; current ESDF map refers to the ESDF map corresponding to the autonomous vehicle at the current moment.

[0023] More specifically, the FAST-LIO (Fast LiDAR-Inertial Odometry) algorithm can be used to fuse the current radar point cloud data and the current pose inertial data to obtain the 6DoF pose data and static point cloud data of the autonomous vehicle at the current moment. The reliability of each of the 6DoF pose data, current pose inertial data, current satellite positioning data, and current wheel speed information is obtained, and based on the reliability of each of these data, the data fusion weights for each of the following are determined: The data fusion weights corresponding to the pose inertial data, current satellite positioning data, and current wheel speed information are used to perform data fusion processing on the 6DoF pose data, current pose inertial data, current satellite positioning data, and current wheel speed information to obtain the target state data of the autonomous vehicle at the current moment. Based on the 6DoF pose data and static point cloud data, the current grid map corresponding to the autonomous vehicle at the current moment is determined based on the ray projection algorithm. The historical grid map corresponding to the autonomous vehicle at the previous moment is obtained, and based on the historical grid map and the current grid map, the current ESDF map is generated based on the signed distance field generation algorithm, such as the FIESTA algorithm.

[0024] Next, the current location coordinates, destination location coordinates, and current ESDF map from the target state data are input into a preset path planning algorithm, such as the Topological Probabilistic Roadmap (TopoPRM) algorithm, to generate multiple collision-free candidate paths; based on the following A-star algorithm's comprehensive cost function: ; From the multiple collision-free candidate paths obtained, one path to be optimized is selected. Using the B-spline curve optimizer, the objective function is to minimize the square integral of the fourth derivative of the trajectory, Snap, and the constraints are the maximum speed, maximum acceleration, and maximum angular velocity of the autonomous vehicle. The path to be optimized is then processed to obtain the target driving path for the autonomous vehicle to travel from its current position to its destination.

[0025] Where n represents the nth candidate path node after the current position; Represents the comprehensive cost function of node n; This represents the actual path cost from the current location to node n; This represents a heuristic function used to evaluate the heuristically estimated cost from node n to the destination; This represents the weight of the heuristic function.

[0026] in, The weights of the heuristic function are determined based on the signed Euclidean distance of the current location in the current ESDF map, the remaining distance from the current location to the destination, and the local obstacle density of the current location. Specifically, the weights are obtained by weighted summation of these factors. .

[0027] The local obstacle density at the current location is determined as follows: A target grid region is defined in the current ESDF map, centered on the current location. The number of grid cells occupied by obstacles within the target grid region and the total number of grid cells in the target grid region are obtained. Based on the number of grid cells occupied by obstacles and the total number of grid cells, the local obstacle density at the current location is determined. The target grid region refers to the grid region in the current ESDF map used to calculate the local obstacle density at the current location. The number of grid cells occupied by obstacles refers to the number of grid cells within the target grid region that are occupied by obstacles.

[0028] More specifically, in the current ESDF map, with the current location as the center and a preset distance as the radius, the target grid area is determined, and the number of grids occupied by obstacles in the target grid area and the total number of grids in the target grid area are obtained; the ratio between the number of grids occupied by obstacles and the total number of grids is used as the local obstacle density at the current location.

[0029] Understandably, after selecting a path to be optimized from multiple collision-free candidate paths using the A* algorithm, the B-spline curve optimizer is used to optimize the path to be optimized, resulting in the target driving path for the autonomous vehicle to travel from its current location to its destination. This improves the smoothness of the target driving path, ensures that each planned trajectory point in the target driving path meets the vehicle dynamics limit, and reduces tracking errors and control difficulty.

[0030] S103. Obtain the current desired position coordinates of the autonomous vehicle on the target driving path, and determine the longitudinal driving force of the autonomous vehicle at the current moment based on the current desired position coordinates and target state data and the multi-loop PID controller for vehicle longitudinal motion adjustment.

[0031] The desired current position coordinates refer to the position coordinates that the autonomous vehicle is expected to reach at the current moment. The multi-loop PID controller includes a position loop PID controller, a speed loop PID controller, and an acceleration loop PID controller. The position loop PID controller controls the speed of the autonomous vehicle; the speed loop PID controller controls the acceleration of the autonomous vehicle; and the acceleration loop PID controller controls the longitudinal driving force of the autonomous vehicle. The longitudinal driving force of the vehicle refers to the force that drives the wheels to rotate and propels the vehicle forward along the longitudinal direction of travel.

[0032] It should be noted that the position loop PID controller, speed loop PID controller, and acceleration loop PID controller each have their own proportional, integral, and derivative coefficients. Optionally, the proportional, integral, and derivative coefficients of each of these controllers can be dynamically adjusted based on the real-time vehicle speed, real-time steering angle, actual load, real-time road friction coefficient, and real-time slope of the autonomous vehicle.

[0033] For example, when the autonomous vehicle's speed is in the high-speed range, the proportional coefficient of the position loop PID controller is automatically reduced by a first preset ratio to prevent high-frequency oscillations. When the autonomous vehicle's speed is in the low-speed range, the proportional coefficient of the position loop PID controller is automatically increased by a second preset ratio to ensure fine-tuning accuracy. When the actual load of the autonomous vehicle exceeds the set load threshold, the proportional and derivative coefficients of both the speed loop PID controller and the acceleration loop PID controller are automatically amplified by a third preset ratio to overcome the response delay caused by large inertia. When the autonomous vehicle is going uphill, the integral coefficient of the speed loop PID controller is automatically increased by a fourth preset ratio to forcibly increase the driving torque of the autonomous vehicle and push it uphill. When the autonomous vehicle enters a preset low-friction surface (such as muddy grass or slippery ground), the proportional coefficient of the acceleration loop PID controller is automatically reduced by a fifth preset ratio, and the maximum output torque limit of the motor is significantly reduced to achieve anti-slip flexible output similar to a car's TCS (Traction Control System).

[0034] Specifically, using the current moment as an index, the desired current position coordinates of the autonomous vehicle on the target driving path are obtained. Based on the desired longitudinal coordinates in the desired current position coordinates, and the actual longitudinal coordinates and actual longitudinal speed in the target state data, the longitudinal driving force of the autonomous vehicle at the current moment is determined using a multi-loop PID controller for vehicle longitudinal motion adjustment. Here, the desired longitudinal coordinates refer to the x-axis coordinates in the desired current position coordinates; the actual current position coordinates refer to the coordinates of the autonomous vehicle's actual location at the current moment; the actual longitudinal coordinates refer to the x-axis coordinates in the actual current position coordinates; and the actual longitudinal speed refers to the actual speed reached by the autonomous vehicle in its forward and backward travel direction at the current moment.

[0035] S104. Based on the target state data, the current yaw rate, the current desired position coordinates, and the current desired heading angle corresponding to the current desired position coordinates in the current pose inertial data, determine the vehicle's desired yaw moment at the current moment based on the MPC prediction model.

[0036] Here, the current yaw rate refers to the yaw rate of the autonomous vehicle at the current moment. The MPC prediction model is a dynamic model based on model predictive control, used to predict the future trends of vehicle position, speed, and attitude based on the vehicle's current motion state, and to solve for the optimal control input by considering constraints. The desired yaw moment of the entire vehicle refers to the torque required to cause the autonomous vehicle to yaw around its vertical axis, achieving steering adjustment.

[0037] Specifically, based on the current actual lateral coordinates and current actual heading angle in the target state data, the current yaw rate in the current pose inertial data, the current expected lateral coordinates in the current expected position coordinates, and the corresponding current expected heading angle, the expected yaw moment of the autonomous vehicle at the current moment is determined using the MPC prediction model. Here, the current actual lateral coordinates refer to the y-axis coordinates in the current actual position coordinates; the current actual heading angle refers to the actual heading angle of the autonomous vehicle at the current moment; and the current expected lateral coordinates refer to the y-axis coordinates in the current expected position coordinates.

[0038] S105. Determine the motor torque corresponding to each drive wheel of the unmanned vehicle based on the longitudinal driving force and the expected yaw torque of the vehicle.

[0039] Here, motor torque refers to the torsional torque required to be output by the motor of the drive wheel. Specifically, the longitudinal driving force and the desired yaw torque of the entire vehicle are input into a preset torque distribution optimizer. Based on the constraints set by the preset torque distribution optimizer, and with the goal of minimizing energy consumption, the actual longitudinal force of each drive wheel of the autonomous vehicle is output by solving a quadratic programming problem. The effective rolling radius of the drive wheels of the autonomous vehicle is obtained, and based on the effective rolling radius and the actual longitudinal force of each drive wheel, the motor torque corresponding to each drive wheel of the autonomous vehicle is determined according to the following torque determination formula: ; in, This represents the motor torque corresponding to the i-th (i=1,2,3,4) drive wheel in an autonomous vehicle; Let represent the actual longitudinal force on the i-th drive wheel of the autonomous vehicle; r represents the effective rolling radius of the drive wheel. It should be noted that the effective rolling radius of all drive wheels in an autonomous vehicle is the same. Specifically, for a single drive wheel of an autonomous vehicle, the actual longitudinal force refers to the force driving that drive wheel along the forward and backward direction of the autonomous vehicle's travel.

[0040] The preset constraints set by the torque distribution optimizer include at least one of the following: 1) The longitudinal driving force and the desired yaw moment of the vehicle satisfy the following dynamic distribution matrix: ; in, Indicates the longitudinal driving force of the entire vehicle; B represents the desired yaw moment of the entire vehicle; B represents the wheelbase of the autonomous vehicle, which is the distance between the center lines of the left and right wheels on the same axle of the autonomous vehicle. This refers to the first drive wheel in an autonomous vehicle (i.e., Figure 2 The actual longitudinal force of wheel number 1 in the middle; This refers to the second drive wheel in an autonomous vehicle (i.e., Figure 2 The actual longitudinal force of wheel number 3 in the middle; This refers to the third drive wheel in an autonomous vehicle (i.e., Figure 2 The actual longitudinal force of wheel number 4 in the middle; This refers to the fourth drive wheel in an autonomous vehicle (i.e., Figure 2 The actual longitudinal force of wheel #6 in the picture; 2) The resultant force of the actual longitudinal force and the actual lateral force of a single drive wheel of an autonomous vehicle shall not exceed the limit friction force corresponding to the current road surface adhesion conditions; where the actual lateral force refers to the force exerted on the drive wheel along the left and right lateral directions of the autonomous vehicle to maintain the vehicle's steering stability. 3) Based on the vehicle's attitude, inertial parameters, and overall geometric parameters, the vertical loads on each wheel of the autonomous vehicle must conform to the mechanical equilibrium range according to the rigid body mechanics model. 4) The armature current of the motors of each drive wheel of the autonomous vehicle must not exceed the peak limit set by the hardware to avoid burning out the motor.

[0041] Optionally, after obtaining the motor torque corresponding to each drive wheel of the autonomous vehicle, in order to distribute the electrical load of the drive wheels on the same side of the autonomous vehicle, and to extend the life of the motors of each drive wheel and reduce single-point overheating, the following adjustments can be made: The motor torque corresponding to the left front drive wheel and the left rear drive wheel of the autonomous vehicle can be adjusted based on the actual current of the motors corresponding to the left front drive wheel and the left rear drive wheel; simultaneously, the motor torque corresponding to the right front drive wheel and the right rear drive wheel of the autonomous vehicle can be adjusted based on the actual current of the motors corresponding to the right front drive wheel and the right rear drive wheel.

[0042] S106. Control the wheels of the autonomous vehicle according to the motor torque corresponding to each drive wheel of the autonomous vehicle.

[0043] Specifically, for each drive wheel of the autonomous vehicle, a motor control command is generated based on the motor torque corresponding to that drive wheel, and the motor of that drive wheel is rotated according to the motor control command, thereby achieving control of that drive wheel.

[0044] The technical solution of this invention fuses multi-source perception data of the autonomous vehicle at the current moment to obtain the target state data of the autonomous vehicle at the current moment, thereby improving the accuracy and reliability of the target state data. Simultaneously, based on the multi-source perception data of the autonomous vehicle at the current moment, a target driving path is generated, improving the accuracy of the target driving path. Then, based on the target state data and the expected vehicle state data corresponding to each planned trajectory point on the target driving path, combined with a multi-loop PID controller and MPC prediction model for longitudinal motion adjustment, wheel-to-wheel coordination control of the autonomous vehicle is performed. The multi-loop PID controller quickly corrects operational deviations, eliminates steady-state deviations, and suppresses control oscillations, effectively resolving the problem of inter-wheel drive force interference and avoiding inter-wheel conflicts, thus improving the multi-wheel coordination effect of the autonomous vehicle and avoiding risks of exceeding speed, torque, and attitude limits. Furthermore, by using the MPC prediction model to predict the vehicle's driving state in advance, the expected yaw torque of the autonomous vehicle at the current moment is optimized under multiple constraints, effectively suppressing steering sideslip, thereby improving the sideslip suppression effect during differential steering of the autonomous vehicle and enhancing the driving stability and steering safety of the autonomous vehicle. The entire wheel control logic is streamlined and efficient, and can be well adapted to the high real-time vehicle control requirements, ensuring real-time control response of the wheels.

[0045] Example 2 Figure 3 This is a flowchart of a wheel control method for an unmanned vehicle provided in Embodiment 2 of the present invention. Based on the above embodiments, this embodiment further optimizes the step of "determining the longitudinal driving force of the unmanned vehicle at the current moment based on the multi-loop PID controller for adjusting the vehicle's longitudinal motion, according to the current desired position coordinates and target state data," providing an optional implementation scheme. It should be noted that parts not detailed in this embodiment can be referred to in other embodiments. Figure 3 As shown, the method includes: S201. Obtain the multi-source perception data of the autonomous vehicle at the current moment and the destination coordinates of the autonomous vehicle; the multi-source perception data includes the current pose inertial data.

[0046] S202. Based on multi-source perception data and destination location coordinates, determine the target state data of the autonomous vehicle at the current moment, as well as the target driving path from the current location to the destination; the target state data includes the current actual position coordinates, the current actual heading angle, and the current three-axis velocity; the current actual position coordinates include the current lateral actual coordinates, the current longitudinal actual coordinates, and the current spatial vertical direction actual coordinates; the current three-axis velocity includes the current x-axis velocity, the current y-axis velocity, and the current z-axis velocity.

[0047] Among them, the current x-axis speed refers to the speed actually reached by the autonomous vehicle in the forward and backward direction at the current moment; the current y-axis speed refers to the speed actually reached by the autonomous vehicle in the left and right lateral direction at the current moment; and the current z-axis speed refers to the speed actually reached by the autonomous vehicle in the vertical direction in space at the current moment.

[0048] It should be noted that the target driving path is essentially a parameterized state sequence with timestamps, and its form is as follows: .in, This indicates that at time t, the autonomous vehicle is expected to reach the following state: located at position (x, y) with a heading angle of θ. The linear velocity is v, and the angular velocity is... .

[0049] S203. Obtain the current expected position coordinates of the unmanned vehicle on the target driving path; the current expected position coordinates include the current lateral expected coordinates and the current longitudinal expected coordinates.

[0050] S204. Based on the current expected longitudinal coordinates and the current actual longitudinal coordinates, determine the longitudinal mileage error of the autonomous vehicle at the current moment.

[0051] Among them, longitudinal mileage error refers to the error between the current actual longitudinal coordinate and the current expected longitudinal coordinate in the forward and backward travel direction of the autonomous vehicle.

[0052] Specifically, the difference between the current expected longitudinal coordinate and the current actual longitudinal coordinate is taken as the longitudinal mileage error of the autonomous vehicle at the current moment.

[0053] S205. Input the longitudinal mileage deviation into the position loop PID controller to obtain the current longitudinal desired speed of the autonomous vehicle.

[0054] The current longitudinal expected speed refers to the speed that the autonomous vehicle should reach at the current moment in its forward and backward travel direction. It should be noted that the current longitudinal expected speed is limited by the upper limit of the speed output of the position loop PID controller; that is, the current longitudinal expected speed is less than or equal to the upper limit of the speed output of the position loop PID controller.

[0055] The upper limit of the speed output of the position loop PID controller is determined as follows: the desired yaw torque of the vehicle and the current x-axis speed are input into the cross decoupling compensator to obtain the speed attenuation coefficient; the upper limit of the speed output is determined based on the speed attenuation coefficient and the preset speed threshold.

[0056] The preset speed threshold can be pre-set according to actual business needs or the expert experience of those skilled in the art; this embodiment of the invention does not impose specific limitations on it. The speed attenuation coefficient is a coefficient used to limit the upper limit of the speed output of the position loop PID controller, that is, a coefficient used to limit the current desired longitudinal speed.

[0057] Specifically, the upper limit of the speed output is determined based on the speed attenuation coefficient and the preset speed threshold. This can be achieved by using the following formula to determine the upper limit of the speed output of the position loop PID controller: ; in, Indicates the preset speed threshold; Indicates the velocity attenuation coefficient; This indicates the upper limit of the speed output of the position loop PID controller.

[0058] Understandably, by dynamically adjusting the speed output limit of the position loop PID controller based on the vehicle's desired yaw moment and the current x-axis speed, the autonomous vehicle can automatically reduce its longitudinal acceleration capability under large steering demands. This reduces the competition between longitudinal acceleration and lateral steering for tire adhesion, thereby improving the stability and safety of the autonomous vehicle during turning, obstacle avoidance, and differential steering.

[0059] S206. Based on the current x-axis velocity and the current longitudinal desired velocity, determine the longitudinal velocity error of the unmanned vehicle at the current moment.

[0060] Among them, longitudinal velocity error refers to the error between the current x-axis velocity and the current longitudinal desired velocity in the forward and backward travel direction of the autonomous vehicle.

[0061] Specifically, the difference between the current expected longitudinal velocity and the current x-axis velocity is taken as the longitudinal velocity error of the autonomous vehicle at the current moment.

[0062] S207. Input the longitudinal velocity error into the speed loop PID controller to obtain the current desired longitudinal acceleration of the unmanned vehicle.

[0063] The current longitudinal expected acceleration refers to the acceleration that the autonomous vehicle should achieve at the current moment in its forward and backward travel directions. It should be noted that the current longitudinal expected acceleration is limited by the autonomous vehicle's safe cornering speed; that is, the current longitudinal expected acceleration is less than or equal to the first derivative of the vehicle's safe cornering speed.

[0064] The safe cornering speed refers to the speed at which an autonomous vehicle safely passes through a curve. Optionally, the safe cornering speed is determined as follows: the current road surface adhesion coefficient of the autonomous vehicle is obtained, and the current maximum lateral acceleration of the autonomous vehicle is determined based on the current road surface adhesion coefficient; the maximum path curvature within the prediction time window is determined based on the turning radius at each planned trajectory point on the target driving path within the prediction time window; and the safe cornering speed is determined based on the current maximum lateral acceleration and the maximum path curvature.

[0065] The prediction time window refers to a window used to predict the vehicle's state over a period of time after the current moment. Its duration can be dynamically adjusted based on the autonomous vehicle's real-time speed and trajectory complexity. For each trajectory point on the target path, the turning radius at that point is the radius of the circumcircle of the tangent plane at that point. The current maximum lateral acceleration refers to the maximum acceleration that the autonomous vehicle can achieve at the current moment in the left and right lateral directions.

[0066] Specifically, the maximum lateral acceleration of the autonomous vehicle is determined based on the current road surface adhesion coefficient. This can be achieved by multiplying the current road surface adhesion coefficient by the gravitational acceleration, which is then used as the maximum lateral acceleration of the autonomous vehicle.

[0067] Specifically, the maximum path curvature within the prediction time window is determined based on the turning radii of each planned trajectory point on the target driving path within the prediction time window. This can be achieved by: if there are three planned trajectory points on the target driving path that fall within the prediction time window: planned trajectory point 1, planned trajectory point 2, and planned trajectory point 3, and it is known that: 1) the turning radius at planned trajectory point 1 is... The path curvature at trajectory point 1 is The turning radius at trajectory point 2 is... The path curvature at trajectory point 2 is The turning radius at trajectory point 3 is... The path curvature at trajectory point 3 is ;2) Therefore, the maximum path curvature within the prediction time window is the path curvature at the planned trajectory point 1, i.e. .

[0068] Specifically, the safe cornering speed of the vehicle is determined based on the current maximum lateral acceleration and maximum path curvature. This can be achieved using the following formula: ; in, This indicates the safe cornering speed for autonomous vehicles. This indicates the current maximum lateral acceleration; This represents the maximum path curvature.

[0069] Understandably, by constraining the current longitudinal desired acceleration output by the speed loop PID controller using the vehicle's safe cornering speed, the risk of speeding on curves is avoided, ensuring the smoothness and safety of the autonomous vehicle's steering.

[0070] S208. Based on the current x-axis velocity and the current longitudinal desired acceleration, determine the longitudinal acceleration error of the unmanned vehicle at the current moment.

[0071] Here, the current actual longitudinal acceleration refers to the actual acceleration achieved by the autonomous vehicle in its forward and backward travel direction at the current moment. The longitudinal acceleration error refers to the error between the current actual longitudinal acceleration and the current expected longitudinal acceleration in the forward and backward travel direction.

[0072] Specifically, the first derivative of the current x-axis velocity is taken to obtain the current actual longitudinal acceleration; the difference between the current expected longitudinal acceleration and the current actual longitudinal acceleration is taken as the longitudinal acceleration error of the autonomous vehicle at the current moment.

[0073] S209. Input the longitudinal acceleration error into the acceleration loop PID controller to obtain the longitudinal driving force of the unmanned vehicle at the current moment.

[0074] S210. Based on the target state data, the current yaw rate, the current desired position coordinates, and the current desired heading angle corresponding to the current desired position coordinates in the current pose inertial data, determine the vehicle's desired yaw moment at the current moment based on the MPC prediction model.

[0075] Specifically, based on the current actual lateral coordinates and the current expected lateral coordinates, the lateral mileage error of the autonomous vehicle at the current moment is determined; based on the current actual heading angle and the current expected heading angle corresponding to the current expected position coordinates, the heading angle error of the autonomous vehicle at the current moment is determined; the lateral mileage error, heading angle error, and the current yaw rate in the current pose inertial data are input into the MPC prediction model to obtain the vehicle's expected yaw moment at the current moment.

[0076] Among them, lateral distance error refers to the error between the current actual lateral coordinates and the current desired lateral coordinates in the left and right lateral directions of the autonomous vehicle. Heading angle error refers to the error between the current actual heading angle and the current desired heading angle corresponding to the current desired position coordinates.

[0077] More specifically, the difference between the current expected lateral coordinates and the current actual lateral coordinates is taken as the lateral mileage error of the autonomous vehicle at the current moment; the difference between the current expected heading angle corresponding to the current expected position coordinates and the current actual heading angle is taken as the heading angle error of the autonomous vehicle at the current moment; the lateral mileage error, heading angle error and the current yaw rate in the current pose inertial data are input into the MPC prediction model to obtain the vehicle's expected yaw moment at the current moment.

[0078] In the MPC prediction model, the vehicle state variables of the autonomous vehicle at the current moment are defined as follows: ; Where k represents the current time; This represents the vehicle state of the autonomous vehicle at the current moment; This represents the lateral distance error of the autonomous vehicle at the current moment. The heading angle error of the autonomous vehicle at the current moment; This indicates the current yaw rate of the autonomous vehicle.

[0079] The control input of the MPC prediction model at the current moment is defined as follows: ;in, This represents the desired yaw moment of the entire vehicle that needs to be solved at the current moment; This represents the control input that the MPC prediction model needs to obtain through optimization at the current moment.

[0080] The discrete state equation of the MPC prediction model is as follows: ;in, This represents the vehicle state of the autonomous vehicle at the current moment and the next moment. Represents the time-varying state transition matrix of the vehicle; This represents the time-varying control input matrix for the vehicle. It should be noted that... and The system updates parameters in real time based on the autonomous vehicle's current actual speed, current expected speed, current expected acceleration, yaw moment of inertia, wheelbase, tire adhesion characteristics, and motor output performance. Yaw moment of inertia is a measure of the autonomous vehicle's inertia when rotating about the vertical z-axis, used to determine the ease of steering.

[0081] Within the preset prediction time domain, based on the vehicle state variables of the autonomous vehicle at the current moment, the future state sequence obtained through the MPC prediction model is as follows: ; in, Indicates the length of the preset prediction time domain; This represents the prediction of the autonomous vehicle's state at time k+1 based on the vehicle's state at the current time. Then, based on the actual and expected vehicle state at time k, and the preset change in the control input vector within the prediction time domain, the objective function of the MPC prediction model is constructed as follows: ; in, This represents the objective function of the MPC prediction model; This indicates the length of the preset prediction time domain, i.e., how many steps the MPC prediction model predicts into the future; This indicates the length of the control time domain, i.e., how many steps the MPC prediction model needs to optimize the control input vector. This indicates that the autonomous vehicle's state at time k+i is predicted based on the vehicle state at the current time. This represents the expected vehicle state quantity of the autonomous vehicle on the target driving path at time k+i. This represents the state tracking error weight matrix, used to adjust the importance of lateral odometer error, heading angle error, and yaw rate error. This represents the weight matrix of the control input, used to constrain the sudden change in the expected yaw moment of the whole vehicle and improve the steering smoothness of autonomous vehicles; This indicates that the control input of the autonomous vehicle at time k+1 is predicted based on the control input of the autonomous vehicle at the current time. This represents the change between the expected yaw moment of the vehicle at time k+i and the expected yaw moment of the vehicle at time k.

[0082] It should be noted that in the above objective function The term is used to reduce the trajectory tracking error of autonomous vehicles, making them as close as possible to the desired trajectory; in the objective function of the MPC prediction model... This item is used to limit sudden changes in control input, making the change in the desired yaw torque of the whole vehicle smoother, so as to avoid the vehicle turning sharply, slipping or attitude oscillation.

[0083] After obtaining the objective function of the MPC prediction model, within the prediction time domain, based on the constraints of the MPC prediction model, the SQP algorithm is used to continuously adjust the control input sequence to minimize the objective function, thereby obtaining the optimal control input sequence. The first control input in the optimal control input sequence, i.e. , which represents the expected yaw moment of the autonomous vehicle at the current moment.

[0084] The constraints of the MPC prediction model include at least one of the following: 1) The lateral slip velocity at the axis (i.e., the intermediate axis) where the two support wheels of the unmanned vehicle are located is approximately 0.

[0085] 2) The actual longitudinal forces on the four drive wheels of the autonomous vehicle satisfy the following dynamic distribution matrix: ; in, Indicates the longitudinal driving force of the entire vehicle; B represents the desired yaw moment of the entire vehicle; B represents the wheelbase of the autonomous vehicle, which is the distance between the center lines of the left and right wheels on the same axle of the autonomous vehicle. This refers to the first drive wheel in an autonomous vehicle (i.e., Figure 2 The actual longitudinal force of wheel number 1 in the middle; This refers to the second drive wheel in an autonomous vehicle (i.e., Figure 2 The actual longitudinal force of wheel number 3 in the middle; This refers to the third drive wheel in an autonomous vehicle (i.e., Figure 2 The actual longitudinal force of wheel number 4 in the middle; This refers to the fourth drive wheel in an autonomous vehicle (i.e., Figure 2 The actual longitudinal force of wheel number 6 in the model.

[0086] 3) The yaw rate of the unmanned vehicle is within the extreme range limited by chassis speed, tire adhesion characteristics and overall vehicle stability.

[0087] 4) The yaw acceleration of the unmanned vehicle shall not exceed the allowable range corresponding to the tire grip limit and the motor torque limit.

[0088] 5) The expected yaw torque of the whole vehicle output by the MPC prediction model is constrained within the executable range of the motor output capacity and the tire adhesion limit.

[0089] S211. Determine the motor torque corresponding to each drive wheel of the unmanned vehicle based on the longitudinal driving force of the whole vehicle and the expected yaw torque of the whole vehicle.

[0090] S212. Control the wheels of the autonomous vehicle according to the motor torque corresponding to each drive wheel of the autonomous vehicle.

[0091] The technical solution of this invention fuses multi-source perception data of the autonomous vehicle at the current moment to obtain the target state data of the autonomous vehicle at the current moment, thereby improving the accuracy and reliability of the target state data. Simultaneously, based on the multi-source perception data of the autonomous vehicle at the current moment, a target driving path is generated, improving the accuracy of the target driving path. Then, based on the target state data and the expected vehicle state data corresponding to each planned trajectory point on the target driving path, combined with a multi-loop PID controller and MPC prediction model for longitudinal motion adjustment, wheel-to-wheel coordination control of the autonomous vehicle is performed. The multi-loop PID controller quickly corrects operational deviations, eliminates steady-state deviations, and suppresses control oscillations, effectively resolving the problem of inter-wheel drive force interference and avoiding inter-wheel conflicts, thus improving the multi-wheel coordination effect of the autonomous vehicle and avoiding risks of exceeding speed, torque, and attitude limits. Furthermore, by using the MPC prediction model to predict the vehicle's driving state in advance, the expected yaw torque of the autonomous vehicle at the current moment is optimized under multiple constraints, effectively suppressing steering sideslip, thereby improving the sideslip suppression effect during differential steering of the autonomous vehicle and enhancing the driving stability and steering safety of the autonomous vehicle. The entire wheel control logic is streamlined and efficient, and can be well adapted to the high real-time vehicle control requirements, ensuring real-time control response of the wheels.

[0092] Example 3 Figure 4 This is a schematic diagram of a wheel control device for an unmanned vehicle according to Embodiment 3 of the present invention. This embodiment is applicable to situations where the wheels of an unmanned vehicle are controlled. The device can be implemented in hardware and / or software and can be configured in an electronic device. Figure 4 As shown, the device includes: The data acquisition module 301 is used to acquire the multi-source perception data of the autonomous vehicle at the current moment and the destination coordinates of the autonomous vehicle; the multi-source perception data includes the current pose inertial data; The target driving path determination module 302 is used to determine the target state data of the unmanned vehicle at the current moment and the target driving path from the current position to the destination based on multi-source perception data and destination location coordinates. The vehicle longitudinal driving force determination module 303 is used to obtain the current expected position coordinates of the unmanned vehicle on the target driving path, and determine the vehicle longitudinal driving force at the current moment based on the current expected position coordinates and target state data and the multi-loop PID controller for vehicle longitudinal motion adjustment; the multi-loop PID controller includes a position loop PID controller, a speed loop PID controller and an acceleration loop PID controller. The whole vehicle expected yaw moment determination module 304 is used to determine the whole vehicle expected yaw moment of the autonomous vehicle at the current moment based on the target state data, the current yaw angular velocity in the current pose inertial data, the current expected position coordinates and the expected heading angle corresponding to the current expected position coordinates, and the MPC prediction model. The wheel motor torque determination module 305 is used to determine the motor torque corresponding to each drive wheel of the unmanned vehicle based on the longitudinal driving force of the whole vehicle and the desired yaw torque of the whole vehicle. The wheel control module 306 is used to control the wheels of the autonomous vehicle according to the motor torque corresponding to each drive wheel of the autonomous vehicle.

[0093] The technical solution of this invention involves acquiring multi-source perception data of an autonomous vehicle at the current moment and the destination coordinates of the autonomous vehicle; the multi-source perception data includes current pose inertial data; based on the multi-source perception data and the destination coordinates, the target state data of the autonomous vehicle at the current moment and the target driving path from the current position to the destination are determined; the current desired position coordinates of the autonomous vehicle on the target driving path are acquired, and based on the current desired position coordinates and the target state data, a multi-loop PID controller for adjusting the vehicle's longitudinal motion is used to determine the longitudinal driving force of the autonomous vehicle at the current moment; the multi-loop PID controller includes a position loop PID controller, a velocity loop PID controller, and an acceleration loop PID controller; based on the target state data, the current yaw rate in the current pose inertial data, the current desired position coordinates, and the current desired heading angle corresponding to the current desired position coordinates, the desired yaw torque of the autonomous vehicle at the current moment is determined based on an MPC prediction model; based on the longitudinal driving force and the desired yaw torque, the motor torque corresponding to each drive wheel of the autonomous vehicle is determined; and based on the motor torque corresponding to each drive wheel of the autonomous vehicle, the wheels of the autonomous vehicle are controlled. The aforementioned technical solution, by fusing multi-source perception data of the autonomous vehicle at the current moment, obtains the target state data of the autonomous vehicle at the current moment, improving the accuracy and reliability of the target state data. Simultaneously, based on the multi-source perception data of the autonomous vehicle at the current moment, a target driving path is generated, improving the accuracy of the target driving path. Then, based on the target state data and the expected vehicle state data corresponding to each planned trajectory point on the target driving path, combined with a multi-loop PID controller and MPC prediction model for longitudinal motion adjustment, wheel-to-wheel coordinated control of the autonomous vehicle is performed. The multi-loop PID controller quickly corrects operational deviations, eliminates steady-state deviations, and suppresses control oscillations, effectively resolving the problem of inter-wheel drive force interference and avoiding inter-wheel conflicts, thereby improving the multi-wheel coordination effect of the autonomous vehicle and mitigating the risks of exceeding speed, torque, and attitude limits. Furthermore, by using the MPC prediction model to predict the vehicle's driving state in advance, the expected yaw torque of the autonomous vehicle at the current moment is optimized under multiple constraints, effectively suppressing steering sideslip, thereby improving the sideslip suppression effect during differential steering of the autonomous vehicle and enhancing its driving stability and steering safety. The entire wheel control logic is streamlined and efficient, and can be well adapted to the high real-time vehicle control requirements, ensuring real-time control response of the wheels.

[0094] Optionally, the target status data includes the current actual position coordinates, the current actual heading angle, and the current three-axis velocities; the current actual position coordinates include the current actual lateral coordinates, the current actual longitudinal coordinates, and the current actual coordinates in the vertical direction of space; the current three-axis velocities include the current x-axis velocity, the current y-axis velocity, and the current z-axis velocity; the current desired position coordinates include the current desired lateral coordinates and the current desired longitudinal coordinates. The vehicle longitudinal drive force determination module 303 is specifically used for: Based on the current expected longitudinal coordinates and the current actual longitudinal coordinates, determine the longitudinal mileage error of the autonomous vehicle at the current moment; The longitudinal mileage deviation is input into the position loop PID controller to obtain the current longitudinal desired speed of the autonomous vehicle; the current longitudinal desired speed is limited by the upper limit of the speed output of the position loop PID controller. Based on the current x-axis velocity and the current desired longitudinal velocity, determine the longitudinal velocity error of the autonomous vehicle at the current moment; The longitudinal velocity error is input into the speed loop PID controller to obtain the current longitudinal expected acceleration of the autonomous vehicle; the current longitudinal expected acceleration is limited by the safe cornering speed of the autonomous vehicle. Based on the current x-axis velocity and the current desired longitudinal acceleration, determine the longitudinal acceleration error of the autonomous vehicle at the current moment; The longitudinal acceleration error is input into the acceleration loop PID controller to obtain the longitudinal driving force of the autonomous vehicle at the current moment.

[0095] Optionally, the whole vehicle desired yaw moment determination module 304 is specifically used for: Based on the current actual lateral coordinates and the current expected lateral coordinates, determine the lateral mileage error of the autonomous vehicle at the current moment; Based on the current actual heading angle and the current expected heading angle corresponding to the current expected position coordinates, determine the heading angle error of the unmanned vehicle at the current moment; By inputting the lateral mileage error, heading angle error, and the current yaw rate from the current pose inertial data into the MPC prediction model, the expected yaw moment of the autonomous vehicle at the current moment can be obtained.

[0096] Optionally, the device further includes a speed output upper limit determination module, which is specifically used for: The desired yaw moment of the whole vehicle and the current x-axis speed are input into the cross decoupling compensator to obtain the speed attenuation coefficient; The upper limit of the speed output is determined based on the speed attenuation coefficient and the preset speed threshold.

[0097] Optionally, the device further includes a vehicle safe cornering speed determination module, which is specifically used for: Obtain the current road surface adhesion coefficient of the autonomous vehicle, and determine the current maximum lateral acceleration of the autonomous vehicle based on the current road surface adhesion coefficient; The maximum path curvature within the prediction time window is determined based on the turning radius at each planned trajectory point on the target driving path within the prediction time window. Determine the safe cornering speed for the vehicle based on the current maximum lateral acceleration and maximum path curvature.

[0098] Optionally, the multi-source sensing data may also include current radar point cloud data, current satellite positioning data, and current wheel speed information; The target driving path determination module 302 is specifically used for: Data fusion processing is performed on the current radar point cloud data and the current pose inertial data to obtain the 6DoF pose data and static point cloud data of the unmanned vehicle at the current moment. Data fusion processing is performed on 6DoF pose data, current pose inertial data, current satellite positioning data, and current wheel speed information to obtain the target state data of the unmanned vehicle at the current moment; Based on 6DoF pose data and static point cloud data, the current grid map corresponding to the autonomous vehicle at the current moment is determined using the ray casting algorithm. Obtain the historical grid map corresponding to the autonomous vehicle at the previous moment, and generate the current ESDF map based on the historical grid map and the current grid map using the signed distance field generation algorithm. Based on the target state data, destination location coordinates, and the current ESDF map, the target driving path of the autonomous vehicle from the current location to the destination is determined using the A* algorithm. The weights of the heuristic function in the A* algorithm are determined based on the signed Euclidean distance value of the current location point in the current ESDF map, the remaining distance from the current location point to the destination, and the local obstacle density of the current location point.

[0099] Optionally, the device further includes a local obstacle density determination module, which is specifically used for: In the current ESDF map, with the current location as the center, determine the target grid area, and obtain the number of grids occupied by obstacles in the target grid area and the total number of grids in the target grid area; The local obstacle density at the current location is determined based on the number of grid cells occupied by obstacles and the total number of grid cells.

[0100] The wheel control device for unmanned vehicles provided in the embodiments of the present invention can execute the wheel control method for unmanned vehicles provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects for executing the wheel control method for each unmanned vehicle.

[0101] According to embodiments of the present invention, the present invention also provides an electronic device, a readable storage medium, and a computer program product.

[0102] Example 4 Figure 5 A schematic diagram of an electronic device 10, which can be used to implement embodiments of the present invention, is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0103] like Figure 5 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded into the RAM 13 from storage unit 18. The RAM 13 can also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0104] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0105] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, digital signal processors (DSPs), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as wheel control methods for autonomous vehicles.

[0106] In some embodiments, the wheel control method for an autonomous vehicle may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded into and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the wheel control method for an autonomous vehicle described above may be performed. Alternatively, in other embodiments, processor 11 may be configured to perform the wheel control method for an autonomous vehicle by any other suitable means (e.g., by means of firmware).

[0107] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0108] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0109] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0110] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0111] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0112] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0113] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0114] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A wheel control method for an unmanned vehicle, characterized in that, The unmanned vehicle is equipped with four drive wheels and two support wheels, and the method includes: Acquire multi-source perception data of the autonomous vehicle at the current moment and the destination coordinates of the autonomous vehicle; the multi-source perception data includes the current pose inertial data; Based on the multi-source perception data and the destination location coordinates, the target state data of the autonomous vehicle at the current moment and the target driving path from the current location to the destination are determined. The system obtains the current desired position coordinates of the autonomous vehicle on the target driving path, and determines the longitudinal driving force of the autonomous vehicle at the current moment based on the current desired position coordinates and the target state data, using a multi-loop PID controller for vehicle longitudinal motion adjustment; the multi-loop PID controller includes a position loop PID controller, a speed loop PID controller and an acceleration loop PID controller. Based on the target state data, the current yaw rate in the current pose inertial data, the current expected position coordinates, and the current expected heading angle corresponding to the current expected position coordinates, the expected yaw moment of the unmanned vehicle at the current moment is determined based on the MPC prediction model. Based on the longitudinal driving force of the vehicle and the desired yaw moment of the vehicle, determine the motor torque corresponding to each drive wheel of the unmanned vehicle; The wheels of the autonomous vehicle are controlled according to the motor torque corresponding to each drive wheel.

2. The method according to claim 1, characterized in that, The target state data includes the current actual position coordinates, the current actual heading angle, and the current three-axis velocities; the current actual position coordinates include the current actual lateral coordinates, the current actual longitudinal coordinates, and the current actual coordinates in the vertical direction of space; the current three-axis velocities include the current x-axis velocity, the current y-axis velocity, and the current z-axis velocity; the current desired position coordinates include the current desired lateral coordinates and the current desired longitudinal coordinates. Based on the current desired position coordinates and the target state data, and using a multi-loop PID controller for vehicle longitudinal motion adjustment, the longitudinal driving force of the autonomous vehicle at the current moment is determined, including: Based on the current expected longitudinal coordinates and the current actual longitudinal coordinates, determine the longitudinal mileage error of the autonomous vehicle at the current moment; The longitudinal mileage deviation is input into the position loop PID controller to obtain the current longitudinal desired speed of the autonomous vehicle; the current longitudinal desired speed is limited by the speed output upper limit of the position loop PID controller. Based on the current x-axis velocity and the current longitudinal expected velocity, determine the longitudinal velocity error of the autonomous vehicle at the current moment; The longitudinal speed error is input into the speed loop PID controller to obtain the current longitudinal expected acceleration of the autonomous vehicle; the current longitudinal expected acceleration is limited by the safe cornering speed of the autonomous vehicle. Based on the current x-axis velocity and the current longitudinal expected acceleration, determine the longitudinal acceleration error of the autonomous vehicle at the current moment; The longitudinal acceleration error is input into the acceleration loop PID controller to obtain the longitudinal driving force of the unmanned vehicle at the current moment.

3. The method according to claim 2, characterized in that, Based on the target state data, the current yaw rate in the current pose inertial data, the current desired position coordinates, and the current desired heading angle corresponding to the current desired position coordinates, the expected yaw moment of the autonomous vehicle at the current moment is determined using the MPC prediction model, including: Based on the current actual lateral coordinates and the current expected lateral coordinates, determine the lateral mileage error of the autonomous vehicle at the current moment; Based on the current actual heading angle and the current expected heading angle corresponding to the current expected position coordinates, the heading angle error of the unmanned vehicle at the current moment is determined; The lateral mileage error, the heading angle error, and the current yaw rate from the current pose inertial data are input into the MPC prediction model to obtain the expected yaw moment of the unmanned vehicle at the current moment.

4. The method according to claim 3, characterized in that, The upper limit of the speed output is determined in the following way: The desired yaw moment of the vehicle and the current x-axis speed are input into the cross decoupling compensator to obtain the speed attenuation coefficient; The upper limit of the speed output is determined based on the speed attenuation coefficient and the preset speed threshold.

5. The method according to claim 2, characterized in that, The safe cornering speed of the vehicle is determined in the following way: Obtain the current road surface adhesion coefficient of the autonomous vehicle, and determine the current maximum lateral acceleration of the autonomous vehicle based on the current road surface adhesion coefficient; The maximum path curvature within the prediction time window is determined based on the turning radius at each planned trajectory point on the target driving path within the prediction time window. The safe cornering speed of the vehicle is determined based on the current maximum lateral acceleration and the maximum path curvature.

6. The method according to claim 1, characterized in that, The multi-source sensing data also includes current radar point cloud data, current satellite positioning data, and current wheel speed information; The step of determining the target state data of the autonomous vehicle at the current moment and the target driving path from the current location to the destination based on the multi-source sensing data and the destination location coordinates includes: The current radar point cloud data and the current pose inertial data are fused to obtain the 6DoF pose data and static point cloud data of the unmanned vehicle at the current moment. The 6DoF pose data, the current pose inertial data, the current satellite positioning data, and the current wheel speed information are fused to obtain the target state data of the unmanned vehicle at the current moment. Based on the 6DoF pose data and the static point cloud data, the current grid map corresponding to the autonomous vehicle at the current moment is determined using a ray casting algorithm. Obtain the historical grid map corresponding to the autonomous vehicle at the previous moment, and generate the current ESDF map based on the historical grid map and the current grid map using a signed distance field generation algorithm; Based on the target state data, the destination location coordinates, and the current ESDF map, the target driving path of the autonomous vehicle from its current location to its destination is determined using the A* algorithm. The weights of the heuristic function in the A* algorithm are determined based on the signed Euclidean distance value of the current location point in the current ESDF map, the remaining distance from the current location point to the destination, and the local obstacle density of the current location point.

7. The method according to claim 6, characterized in that, The local obstacle density at the current location point is determined in the following way: In the current ESDF map, with the current location point as the center, determine the target grid area, and obtain the number of grids occupied by obstacles in the target grid area and the total number of grids in the target grid area; The local obstacle density at the current location is determined based on the number of grid cells occupied by the obstacles and the total number of grid cells.

8. A wheel control device for an unmanned vehicle, characterized in that, The unmanned vehicle is equipped with four drive wheels and two support wheels, and the device includes: The data acquisition module is used to acquire the multi-source perception data of the autonomous vehicle at the current moment and the destination coordinates of the autonomous vehicle; the multi-source perception data includes the current pose inertial data; The target driving path determination module is used to determine the target state data of the autonomous vehicle at the current moment and the target driving path from the current position to the destination based on the multi-source perception data and the destination location coordinates. The vehicle longitudinal driving force determination module is used to obtain the current desired position coordinates of the autonomous vehicle on the target driving path, and determine the vehicle longitudinal driving force at the current moment based on the current desired position coordinates and the target state data, using a multi-loop PID controller for vehicle longitudinal motion adjustment; the multi-loop PID controller includes a position loop PID controller, a speed loop PID controller and an acceleration loop PID controller. The vehicle expected yaw moment determination module is used to determine the vehicle expected yaw moment at the current moment based on the target state data, the current yaw angular velocity in the current pose inertial data, the current expected position coordinates, and the expected heading angle corresponding to the current expected position coordinates, and the MPC prediction model. The wheel motor torque determination module is used to determine the motor torque corresponding to each drive wheel of the unmanned vehicle based on the longitudinal driving force of the vehicle and the desired yaw torque of the vehicle. The wheel control module is used to control the wheels of the autonomous vehicle according to the motor torque corresponding to each drive wheel of the autonomous vehicle.

9. An electronic device, characterized in that, The electronic device includes: At least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the wheel control method of the unmanned vehicle according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that, when executed by a processor, implement the wheel control method of any one of claims 1-7 for an unmanned vehicle.

Images