Single-rod wheel foot type special vehicle obstacle crossing control method, system and equipment and medium

Abstract

The invention discloses a single-pole wheel-foot type special vehicle obstacle crossing control method, system and device and a medium, and relates to the field of special vehicle obstacle crossing, and the method comprises the steps: building a vehicle dynamics model according to a vehicle control quantity and a vehicle state quantity based on a single-pole wheel-foot type special vehicle obstacle crossing scene; acquiring historical driving data of the single-pole wheel-foot type special vehicle based on the vehicle dynamics model, and training a Gaussian process regression model according to the historical driving data to obtain a trained Gaussian process regression model; and according to the real vehicle state quantity at the current moment, on the basis of the vehicle dynamics model and the trained Gaussian process regression model, vehicle control quantities at all moments in a future set time period are determined through a minimum cost function so as to control the single-pole wheel-foot type special vehicle to complete obstacle crossing. Through the vehicle dynamics model and the Gaussian process regression model, the single-pole wheel foot type special vehicle is controlled to better complete obstacle crossing.

Description

Technical Field

[0001] The present application relates to the field of special vehicle obstacle crossing, and particularly to a control method, system, device and medium for a single-bar wheel-foot special vehicle to cross obstacles. Background Art

[0002] In the field of special vehicle obstacle crossing, single-bar wheel-foot vehicles have broad application prospects due to their advantages such as simple structure and strong mobility. However, current research on the dynamic models of single-bar wheel-foot special vehicles is relatively scarce, especially the lack of targeted dynamic modeling methods in obstacle-crossing scenarios. In addition, during the obstacle-crossing process, multiple factors such as complex terrain conditions, environmental uncertainties, and the approximation of vehicle dynamic models will cause model errors. The model errors lead to deviations between the model and the actual vehicle motion state, making it difficult for traditional control algorithms to accurately handle the vehicle's motion behavior, thereby reducing the control performance and ultimately affecting the maneuverability, obstacle-crossing ability, and control robustness of special vehicles. Summary of the Invention

[0003] The purpose of the present application is to provide a control method, device, equipment and medium for a single-bar wheel-foot special vehicle to cross obstacles, which can control the single-bar wheel-foot special vehicle to better complete obstacle crossing through vehicle dynamic models and Gaussian process regression models.

[0004] To achieve the above purpose, the present application provides the following solutions: In the first aspect, the present application provides a control method for a single-bar wheel-foot special vehicle to cross obstacles, including: Based on the obstacle-crossing scenario of the single-bar wheel-foot special vehicle, a vehicle dynamic model is established according to vehicle control quantities and vehicle state quantities; the vehicle dynamic model is used to predict the vehicle state quantity at the next moment according to the vehicle control quantity at the current moment and the vehicle state quantity at the current moment.

[0005] Historical driving data of the single-bar wheel-foot special vehicle is obtained based on the vehicle dynamic model, and a Gaussian process regression model is trained according to the historical driving data to obtain a trained Gaussian process regression model; the historical driving data includes vehicle control quantities, real vehicle state quantities, and vehicle dynamic model errors at each moment within a historical set period of time; the trained Gaussian process regression model is used to predict the vehicle dynamic model error at the next moment according to the vehicle control quantity at the current moment and the vehicle state quantity at the current moment, so as to correct the vehicle state quantity predicted by the vehicle dynamic model to obtain a predicted vehicle state quantity.

[0006] Based on the true vehicle state variables at the current moment, and based on the vehicle dynamics model and the trained Gaussian process regression model, the vehicle control variables at each moment within a future set time period are determined by minimizing the cost function, so as to control the single-bar wheel-legged special vehicle to complete obstacle crossing; the cost function is the sum of the deviation of the predicted vehicle state variables and the change in the vehicle control variables at each moment within the future set time period; the deviation of the predicted vehicle state variables is the difference between the predicted vehicle state variables and the reference vehicle state variables; the change in the vehicle control variables is the difference between the vehicle control variables at the current moment and the vehicle control variables at the previous moment.

[0007] In a second aspect, the present application provides an intelligent obstacle-crossing control system for a single-bar wheel-legged special vehicle, including: A vehicle dynamics model establishment module, configured to establish a vehicle dynamics model based on the obstacle-crossing scenario of the single-bar wheel-legged special vehicle according to the vehicle control variables and the vehicle state variables; the vehicle dynamics model is used to predict the vehicle state variables at the next moment according to the vehicle control variables at the current moment and the vehicle state variables at the current moment.

[0008] A Gaussian process regression model training module, connected to the vehicle dynamics model establishment module, configured to obtain the historical driving data of the single-bar wheel-legged special vehicle based on the vehicle dynamics model, and train the Gaussian process regression model according to the historical driving data to obtain a trained Gaussian process regression model; the historical driving data includes the vehicle control variables, the true vehicle state variables, and the vehicle dynamics model error at each moment within a historical set time period; the trained Gaussian process regression model is used to predict the vehicle dynamics model error at the next moment according to the vehicle control variables at the current moment and the vehicle state variables at the current moment, so as to correct the vehicle state variables predicted by the vehicle dynamics model to obtain predicted vehicle state variables.

[0009] A vehicle control module, connected to the vehicle dynamics model establishment module and the Gaussian process regression training model, configured to determine the vehicle control variables at each moment within a future set time period based on the true vehicle state variables at the current moment, based on the vehicle dynamics model and the trained Gaussian process regression model, by minimizing the cost function, so as to control the single-bar wheel-legged special vehicle to complete obstacle crossing; the cost function is the sum of the deviation of the predicted vehicle state variables and the change in the vehicle control variables at each moment within the future set time period; the deviation of the predicted vehicle state variables is the difference between the predicted vehicle state variables and the reference vehicle state variables; the change in the vehicle control variables is the difference between the vehicle control variables at the current moment and the vehicle control variables at the previous moment.

[0010] In a third aspect, the present application provides a computer device, including: a memory, a processor, and a computer program stored on the memory and executable on the processor, and the processor executes the computer program to implement the above-mentioned obstacle-crossing control method for a single-bar wheel-legged special vehicle.

[0011] In a fourth aspect, the present application provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the above-mentioned obstacle-crossing control method for a single-rod wheel-legged special vehicle is implemented.

[0012] According to the specific embodiments provided by the present application, the present application has the following technical effects: The present application provides an obstacle-crossing control method, device, equipment and medium for a single-rod wheel-legged special vehicle. First, a vehicle dynamics model of the single-rod wheel-legged special vehicle is established. Secondly, the Gaussian process regression model is used to correct the error generated by the vehicle dynamics model. Finally, the cost function is minimized to determine the vehicle control amounts at each moment within a future set time period, so as to better control the single-rod wheel-legged special vehicle to complete obstacle crossing. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required to be used in the embodiments. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0014] Figure 1 It is a schematic flowchart of an obstacle-crossing control method for a single-rod wheel-legged special vehicle according to the present application.

[0015] Figure 2 It is a schematic diagram of the overall dynamic analysis of a single-rod wheel-legged special vehicle applicable to an obstacle-crossing scenario according to the present application.

[0016] Figure 3 For Figure 2 It is a schematic diagram of the local dynamic analysis of the first single rod of the single-rod wheel-legged special vehicle in

[0017] Figure 4 For Figure 2 It is a schematic diagram of the local dynamic analysis of the second single rod of the single-rod wheel-legged special vehicle in

[0018] Figure 5 For Figure 2 It is a schematic diagram of the local dynamic analysis of the third single rod of the single-rod wheel-legged special vehicle in

[0019] Figure 6 It is an architecture diagram for collecting historical data.

[0020] Figure 7 It is a schematic structural diagram of a computer device provided by an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0021] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

[0022] To make the above objects, features, and advantages of the present application more obvious and understandable, the present application will be further described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0023] In an exemplary embodiment, as Figure 1 shown, a method for obstacle-crossing control of a single-rod wheel-legged special vehicle is provided. This method is executed by a computer device, and specifically can be executed alone by a computer device such as a terminal or a server, or can be jointly executed by a terminal and a server, including the following steps 101 to step 103. Among them: Step 101, based on the obstacle-crossing scenario of the single-rod wheel-legged special vehicle, establish a vehicle dynamics model according to the vehicle control quantity and the vehicle state quantity.

[0024] The vehicle dynamics model is used to predict the vehicle state quantity at the next moment according to the vehicle control quantity at the current moment and the vehicle state quantity at the current moment.

[0025] Step 102, obtain the historical driving data of the single-rod wheel-legged special vehicle based on the vehicle dynamics model, and train a Gaussian process regression model according to the historical driving data to obtain a trained Gaussian process regression model.

[0026] The historical driving data includes the vehicle control quantity, the real vehicle state quantity, and the vehicle dynamics model error at each moment within a historical set time period; the trained Gaussian process regression model is used to predict the vehicle dynamics model error at the next moment according to the vehicle control quantity at the current moment and the vehicle state quantity at the current moment, so as to correct the vehicle state quantity predicted by the vehicle dynamics model to obtain a predicted vehicle state quantity.

[0027] Step 103, according to the real vehicle state quantity at the current moment, based on the vehicle dynamics model and the trained Gaussian process regression model, determine the vehicle control quantity at each moment within a future set time period by minimizing the cost function, so as to control the single-rod wheel-legged special vehicle to complete obstacle crossing.

[0028] The cost function is the sum of the predicted vehicle state quantity deviation and the vehicle control quantity change at each moment within a future set time period; the predicted vehicle state quantity deviation is the difference between the predicted vehicle state quantity and the reference vehicle state quantity; the vehicle control quantity change is the difference between the vehicle control quantity at the current moment and the vehicle control quantity at the previous moment.

[0029] In another exemplary embodiment of the present application, the single-bar wheel-legged special vehicle includes a first single bar, a second single bar, a third single bar, a front wheel-legged unit, and a rear wheel-legged unit. The front wheel-legged unit is connected to the rear wheel-legged unit through the first single bar, the second single bar, and the third single bar.

[0030] The vehicle control quantities include: the tire driving force of the front wheel-legged unit, the tire driving force of the rear wheel-legged unit, the joint driving torque of the front wheel-legged unit, and the joint driving torque of the rear wheel-legged unit; the vehicle state quantities include: the X-axis coordinate value of the center of mass of the single-bar wheel-legged special vehicle in a three-dimensional rectangular coordinate system, the Z-axis coordinate value of the center of mass of the single-bar wheel-legged special vehicle in a three-dimensional rectangular coordinate system, and the pitch angle of the second single bar around the Y-axis in the three-dimensional rectangular coordinate system; the X-axis of the three-dimensional rectangular coordinate system is the forward direction of the single-bar wheel-legged special vehicle, and the Y-axis is in the same plane as the X-axis and perpendicular to each other.

[0031] In another exemplary embodiment of the present application, as Figure 2 shown, a vehicle dynamics model is established based on the single-bar wheel-legged special vehicle. Among them, is the tire driving force of the front wheel-legged unit, is the tire driving force of the rear wheel-legged unit, is the ground support force of the front wheel-legged unit, is the ground support force of the rear wheel-legged unit, is the joint driving torque of the front wheel-legged unit, is the joint driving torque of the rear wheel-legged unit, is the velocity of the center of mass CG of the single-bar wheel-legged special vehicle in the X direction, is the velocity of the center of mass CG of the single-bar wheel-legged special vehicle along the Z-axis, is the pitch angle of the second single bar BC around the Y-axis in the three-dimensional rectangular coordinate system, is the first derivative of.

[0032] As Figure 3 , Figure 4 and Figure 5 shown, a kinetic local analysis is respectively performed on the first single bar AB, the second single bar BC, and the third single bar CD. Combining Newton's second law and Euler's rotation law, the following formulas can be obtained.

[0033] (1) (2) (3) (4) (5) Among them, is A the X-axis coordinate value of the point under the three-dimensional rectangular coordinate system XOYZ, is B the X-axis coordinate value of the point under the three-dimensional rectangular coordinate system XOYZ, is C the X-axis coordinate value of the point under the three-dimensional rectangular coordinate system XOYZ, is D the X-axis coordinate value of the point under the three-dimensional rectangular coordinate system XOYZ, is A the Z-axis coordinate value of the point under the three-dimensional rectangular coordinate system XOYZ, is B the Z-axis coordinate value of the point under the three-dimensional rectangular coordinate system XOYZ, is C the Z-axis coordinate value of the point under the three-dimensional rectangular coordinate system XOYZ, is the acceleration of the centroid CG of the single-bar wheeled-legged special vehicle along the X-axis, is the acceleration of the centroid CG of the single-bar wheeled-legged special vehicle along the Z-axis, is the height of the step, is the tire radius of the front wheeled-foot unit and the rear wheeled-foot unit, is the length of the second single bar BC, is the mass of the second single bar BC, is the moment of inertia of the second single bar BC about the Y-axis.

[0034] By arranging equations (1) and (2), the following formula can be obtained.

[0035] (6) (7) Among them, represents the X-axis coordinate value of point B under the coordinate system of point A, represents the Z-axis coordinate value of point B under the coordinate system of point A, represents the X-axis coordinate value of point D under the coordinate system of point C. From the geometric relationship, , , and can be expressed in the following form.

[0036] (8) (9) (10) (11) Among them, is the length of the first single rod AB, is the length of the third single rod CD, is the X-axis coordinate value of the centroid CG of the single-rod wheeled special vehicle, is the Z-axis coordinate value of the centroid CG of the single-rod wheeled special vehicle. Substituting equations (6) to (11) into equations (4) to (5) gives the following equations (12) and (13).

[0037] (12) (13) Let the control quantity of the dynamic model be , and the state quantity be . After arranging equations (3), (12), and (13), the following equation (14) is established, which is the dynamic model of the single-rod wheeled special vehicle applicable to the obstacle-crossing scenario: (14) Among them, is the X-axis coordinate value of the centroid of the single-rod wheeled special vehicle in the three-dimensional rectangular coordinate system, is the Z-axis coordinate value of the centroid of the single-rod wheeled special vehicle in the three-dimensional rectangular coordinate system, is the pitch angle of the second single rod around the Y-axis in the three-dimensional rectangular coordinate system, is the first derivative of is the first derivative of is the first derivative of is the second derivative of is the second derivative of is the second derivative of is the tire driving force of the front wheel-foot unit, is the tire driving force of the rear wheel-foot unit, is the joint driving torque of the front wheel-foot unit, is the joint driving torque of the rear wheel-foot unit, is the mass of the second single rod, is the length of the second single rod, is the height of the step, is the length of the first single rod, is the length of the third single rod, is the tire radius of the front wheel-foot unit and the rear wheel-foot unit, is the moment of inertia of the second single rod around the Y-axis.

[0038] The dynamic model of the single-bar wheel-legged special vehicle can also be expressed in the following mathematical form: .

[0039] In another exemplary embodiment of the present application, historical driving data of the single-bar wheel-legged special vehicle is obtained based on the vehicle dynamic model, specifically including steps 201 to 203. Among them: Step 201, obtaining the vehicle control quantity and the predicted vehicle state quantity at each moment within the historical set time period based on the vehicle dynamic model.

[0040] Step 202, obtaining the actual vehicle state quantity at each moment within the historical set time period based on the actual dynamics.

[0041] Step 203, determining the vehicle dynamic model error at each moment within the historical set time period according to the predicted vehicle state quantity and the actual vehicle state quantity at each moment within the historical set time period.

[0042] Construct the vehicle control quantity and the actual vehicle state quantity at each moment within the historical set time period into a feature vector, and construct the predicted vehicle state quantity and the actual vehicle state quantity at each moment within the historical set time period into a target vector. As Figure 6 shown, the feature vector is . Among them, is the vehicle control quantity at the th moment, that is, the vehicle control quantity at each moment within the historical set time period, i and N are the serial numbers of the moments; is the actual vehicle state quantity at the th moment, that is, the actual vehicle state quantity at each moment within the historical set time period, is the feature component at the th moment. The target vector is . Among them, is the vehicle dynamic model error at the th moment, that is, the vehicle dynamic model error at each moment within the historical set time period, is the predicted vehicle state quantity at the th moment, that is, the predicted vehicle state quantity at each moment within the historical set time period.

[0043] In another exemplary embodiment of the present application, a Gaussian process regression model is trained according to the historical driving data to obtain a trained Gaussian process regression model, specifically including steps 301 to 302. Among them: Step 301, constructing a Gaussian process regression model based on the squared exponential kernel function and the mean function.

[0044] Step 302: Based on the historical driving data, optimize the hyperparameters in the Gaussian process regression model by maximizing the log marginal likelihood method to obtain a trained Gaussian process regression model.

[0045] The Gaussian process regression model is , where is the mean function; is the covariance function, and generally the squared exponential kernel function is used. is the input point, is another input point, representing the correlation between two input points; the form of the squared exponential kernel function is: (15) where is the signal variance, is the length scale, which controls the decay of the similarity between input points. For the feature vector , the elements of the covariance matrix are calculated by the kernel function: (16) Assume that the observed target vector contains observation noise , where are the noise components at different times, is the noise component at the (N + 1)-th time.

[0046] (17) (18) where is the noise variance, is the noise component at the -th time, is expressed as follows a normal distribution with a mean of 0 and a variance of . For the feature vector and the target vector , the joint distribution can be expressed as: (19) where is the identity matrix. For the new input point , the formulas for the predicted mean and the variance are as follows.

[0047] (20) (21) where For the new point and the covariance vector of the feature vector . The hyperparameters in the model , and are optimized by maximizing the log marginal likelihood.

[0048] (22) In another exemplary embodiment of the present application, the discretized vehicle dynamics model is:[[]] (23) wherein, is the vehicle state quantity at , is the current time, is the vehicle control quantity at , is the state transition function of the vehicle dynamics model, is the model error at (24) wherein, is the mean function; is the covariance function, describing the correlation between input points, and are at different times.

[0049] The predicted vehicle state quantity is obtained by the following formula:[[]] (25) wherein, is the predicted vehicle state quantity at , is the vehicle state quantity at , is the vehicle control quantity at is the state transition function of the vehicle dynamics model, is the vehicle dynamics model error at

[0050] Prediction step of model predictive control:[[]] (26) (27) ....... (28) wherein, It is the predicted vehicle state quantity at the time after error adaptive correction using Gaussian process regression. The predicted vehicle state quantity at the time.

[0051] In another exemplary embodiment of the present application, the expression of the cost function is: (29) Wherein, is the vehicle control quantity at the time, is the vehicle control quantity at the time, is the vehicle control quantity at the time, the k-th moment is the current moment, and the moments from the (k + 1)-th moment to the (k + N - 1)-th moment are the moments within the future set time period. is the predicted vehicle state quantity at the time is the reference vehicle state quantity at the time, is the change in the vehicle control quantity at the time, is the vehicle control quantity at the time, is the vehicle control quantity at the time, is the state weight matrix, is the control input weight matrix.

[0052] Based on the same inventive concept, an obstacle-crossing control system for a single-bar wheeled-foot special vehicle provided by an embodiment of the present application is applied to the above-mentioned obstacle-crossing control method for a single-bar wheeled-foot special vehicle. The intelligent obstacle-crossing control system of the single-bar wheeled-foot special vehicle includes: a vehicle dynamics model establishment module, a Gaussian process regression model training module, and a vehicle control module.

[0053] The vehicle dynamics model establishment module is used to establish a vehicle dynamics model based on the obstacle-crossing scenario of the single-bar wheeled-foot special vehicle according to the vehicle control quantity and the vehicle state quantity; the vehicle dynamics model is used to predict the vehicle state quantity at the next moment according to the vehicle control quantity at the current moment and the vehicle state quantity at the current moment.

[0054] The Gaussian process regression model training module is connected to the vehicle dynamics model establishment module. The Gaussian process regression model training module is used to obtain the historical driving data of the single-bar wheeled-foot special vehicle based on the vehicle dynamics model, and train the Gaussian process regression model according to the historical driving data to obtain a trained Gaussian process regression model.

[0055] The historical driving data includes vehicle control quantities, true vehicle state quantities, and vehicle dynamics model errors at each moment within a historical set period of time; the trained Gaussian process regression model is used to predict the vehicle dynamics model error at the next moment based on the vehicle control quantity and the vehicle state quantity at the current moment, so as to correct the vehicle state quantity predicted by the vehicle dynamics model and obtain a predicted vehicle state quantity.

[0056] The vehicle control module is connected to the vehicle dynamics model establishment module and the Gaussian process regression training model. The vehicle control module is used to determine the vehicle control quantities at each moment within a future set period based on the true vehicle state quantity at the current moment, the vehicle dynamics model, and the trained Gaussian process regression model by minimizing a cost function, so as to control the single-bar wheel-legged special vehicle to complete obstacle crossing.

[0057] The cost function is the sum of the predicted vehicle state quantity deviations and the vehicle control quantity changes at each moment within a future set period; the predicted vehicle state quantity deviation is the difference between the predicted vehicle state quantity and the reference vehicle state quantity; the vehicle control quantity change is the difference between the vehicle control quantity at the current moment and the vehicle control quantity at the previous moment.

[0058] In an exemplary embodiment, a computer device is provided. The computer device can be a server or a terminal, and its internal structure diagram can be as Figure 7 shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, abbreviated as I / O), and a communication interface. Among them, the processor, the memory, and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store the driving data of the single-bar wheel-legged special vehicle. The input / output interface of the computer device is used to exchange information between the processor and external devices. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, it implements a method for controlling the obstacle crossing of a single-bar wheel-legged special vehicle.

[0059] Those skilled in the art can understand, Figure 7The structure shown is only a block diagram of some structures related to the solution of this application, and does not constitute a limitation on the computer device to which the solution of this application is applied. Specifically, the computer device may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements. In an exemplary embodiment, a computer device is provided, including a memory and a processor. A computer program is stored in the memory, and when the processor executes the computer program, the steps in the above method embodiments are implemented.

[0060] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0061] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use, and processing of relevant data need to comply with relevant regulations.

[0062] Those of ordinary skill in the art can understand that all or part of the processes of implementing the methods in the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it may include the processes of the above method embodiments. Among them, any reference to a memory, database, or other medium used in the embodiments provided in this application may include at least one of non-volatile and volatile memories. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory may include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0063] In each of the embodiments provided in the present application, the database involved may include at least one of a relational database and a non-relational database. The non-relational database may include a distributed database based on blockchain, etc., and is not limited thereto. In each of the embodiments provided in the present application, the processor involved may be a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., and is not limited thereto.

[0064] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope described in this specification.

[0065] Specific examples are used in this article to elaborate on the principles and implementation manners of the present application. The description of the above embodiments is only used to help understand the method and its core idea of the present application; at the same time, for those of ordinary skill in the art, according to the idea of the present application, there will be changes in the specific implementation manners and application scopes. In summary, the content of this specification should not be construed as a limitation to the present application.

Claims

1. A single-rod wheel-foot special vehicle obstacle control method, characterized in that: The single-rod wheel-foot special vehicle obstacle crossing control method comprises: Based on the obstacle crossing scenario of a single-pole wheel-foot special vehicle, a vehicle dynamics model is established according to the vehicle control quantity and the vehicle state quantity; the vehicle dynamics model is used to predict the vehicle state quantity at the next moment according to the vehicle control quantity at the current moment and the vehicle state quantity at the current moment; Based on the vehicle dynamics model, historical driving data of the single-pole wheel-foot special vehicle is obtained, and a Gaussian process regression model is trained according to the historical driving data to obtain a trained Gaussian process regression model; the historical driving data includes the vehicle control quantity, the real vehicle state quantity and the vehicle dynamics model error at each moment within a historical set period of time; the trained Gaussian process regression model is used to predict the vehicle dynamics model error at the next moment according to the vehicle control quantity and the vehicle state quantity at the current moment, so as to correct the vehicle state quantity predicted by the vehicle dynamics model to obtain the predicted vehicle state quantity; According to the actual vehicle state at the current moment, based on the vehicle dynamics model and the trained Gaussian process regression model, the vehicle control quantity at each moment in a future set time period is determined by minimizing the cost function to control the single-rod wheel-foot special vehicle to complete obstacle crossing; the cost function is the sum of the predicted vehicle state quantity deviation and the vehicle control quantity change at each moment in the future set time period; the predicted vehicle state quantity deviation is the difference between the predicted vehicle state quantity and the reference vehicle state quantity; the vehicle control quantity change is the difference between the vehicle control quantity at the current moment and the vehicle control quantity at the previous moment.

2. The single-rod wheel-foot special vehicle obstacle control method according to claim 1 is characterized in that: The single-rod wheel-foot type special vehicle comprises a first single rod, a second single rod, a third single rod, a front wheel foot unit and a rear wheel foot unit, wherein the front wheel foot unit is connected to the rear wheel foot unit via the first single rod, the second single rod and the third single rod; The vehicle control quantity includes: the tire driving force of the front wheel foot unit, the tire driving force of the rear wheel foot unit, the joint driving torque of the front wheel foot unit and the joint driving torque of the rear wheel foot unit; the vehicle state quantity includes: the X-axis coordinate value of the center of mass of the single-rod wheel foot type special vehicle in the three-dimensional rectangular coordinate system, the Z-axis coordinate value of the center of mass of the single-rod wheel foot type special vehicle in the three-dimensional rectangular coordinate system and the pitch angle of the second single rod around the Y-axis in the three-dimensional rectangular coordinate system; the X-axis of the three-dimensional rectangular coordinate system is the forward direction of the single-rod wheel foot type special vehicle, and the Y-axis and the X-axis are located on the same plane and are perpendicular to each other.

3. The obstacle-crossing control method for a single-rod wheel-foot special vehicle according to claim 1, characterized in that: The historical driving data of the single-rod wheel-foot special vehicle is obtained based on the vehicle dynamics model, specifically including: Based on the vehicle dynamics model, the vehicle control quantity and the predicted vehicle state quantity at each moment in a historical set period of time are obtained; Based on actual dynamics, the real vehicle state quantity at each moment in the historical set period is obtained; The vehicle dynamics model error at each moment in the historical setting period is determined according to the predicted vehicle state quantity and the actual vehicle state quantity at each moment in the historical setting period.

4. The obstacle-crossing control method for a single-rod wheel-foot special vehicle according to claim 1, characterized in that: The Gaussian process regression model is trained according to the historical driving data to obtain a trained Gaussian process regression model, which specifically includes: A Gaussian process regression model is constructed based on the square exponential kernel function and the mean function; Based on the historical driving data, the hyperparameters in the Gaussian process regression model are optimized by maximizing the logarithmic marginal likelihood method to obtain a trained Gaussian process regression model.

5. The obstacle-crossing control method for a single-rod wheel-foot special vehicle according to claim 1, characterized in that: The vehicle dynamics model is: ； in, is the X-axis coordinate value of the mass center of the single-rod wheel-foot special vehicle in the three-dimensional rectangular coordinate system, is the Z-axis coordinate value of the center of mass of the single-rod wheel-foot special vehicle in the three-dimensional rectangular coordinate system, is the pitch angle of the second rod around the Y axis in the three-dimensional rectangular coordinate system, for The first derivative of for The first derivative of for The first derivative of for The second-order derivative of for The second-order derivative of for The second-order derivative of is the tire driving force of the front wheel foot unit, is the tire driving force of the rear wheel foot unit, is the joint driving torque of the front wheel foot unit, is the joint driving torque of the rear wheel-foot unit, is the mass of the second rod, is the length of the second single rod, is the height of the steps, is the length of the first single rod, is the length of the third single rod, is the tire radius of the front wheel foot unit and the rear wheel foot unit, is the moment of inertia of the second single rod around the Y axis.

6. The obstacle-crossing control method for a single-rod wheel-foot special vehicle according to claim 1, characterized in that: The following formula is used to get the predicted vehicle state: ； in, for The predicted vehicle state at the time, for The vehicle state at the time, The time is the current time, for The amount of vehicle control at the time, is the state transfer function of the vehicle dynamics model, for The vehicle dynamics model error at time t.

7. The obstacle-crossing control method for a single-rod wheel-foot special vehicle according to claim 1, characterized in that: The expression of the cost function is: ； in, for The amount of vehicle control at the time, for The amount of vehicle control at the time, for The vehicle control quantity at the time, time k is the current time, time k+1 to time k+N-1 are the times within the future set period, for The predicted vehicle state at the time, for The reference vehicle state at time for The vehicle control quantity changes at each moment. for The amount of vehicle control at the time, for The amount of vehicle control at the time, is the state weight matrix, is the control input weight matrix, i and N The sequence number of the moment.

8. An intelligent obstacle-crossing control system for a single-rod wheel-footed special vehicle, applied to the obstacle-crossing control method for a single-rod wheel-footed special vehicle according to any one of claims 1 to 7, characterized in that: The intelligent obstacle crossing control system of the single-rod wheel-foot special vehicle comprises: A vehicle dynamics model building module is used to build a vehicle dynamics model based on a single-pole wheel-foot special vehicle obstacle crossing scenario according to a vehicle control quantity and a vehicle state quantity; the vehicle dynamics model is used to predict the vehicle state quantity at the next moment according to the vehicle control quantity at the current moment and the vehicle state quantity at the current moment; A Gaussian process regression model training module is connected to the vehicle dynamics model establishment module, and is used to obtain historical driving data of the single-pole wheel-foot special vehicle based on the vehicle dynamics model, and train the Gaussian process regression model according to the historical driving data to obtain a trained Gaussian process regression model; the historical driving data includes the vehicle control quantity, the real vehicle state quantity and the vehicle dynamics model error at each moment within a historical set period of time; the trained Gaussian process regression model is used to predict the vehicle dynamics model error at the next moment according to the vehicle control quantity and the vehicle state quantity at the current moment, so as to correct the vehicle state quantity predicted by the vehicle dynamics model to obtain a predicted vehicle state quantity; The vehicle control module is connected to the vehicle dynamics model establishment module and the Gaussian process regression training model, and is used to determine the vehicle control quantity at each moment in a future set time period by minimizing the cost function according to the actual vehicle state quantity at the current moment, based on the vehicle dynamics model and the trained Gaussian process regression model, so as to control the single-pole wheel-foot special vehicle to complete obstacle crossing; the cost function is the sum of the predicted vehicle state quantity deviation and the vehicle control quantity change at each moment in the future set time period; the predicted vehicle state quantity deviation is the difference between the predicted vehicle state quantity and the reference vehicle state quantity; the vehicle control quantity change is the difference between the vehicle control quantity at the current moment and the vehicle control quantity at the previous moment.

9. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the single-rod wheel-foot special vehicle obstacle crossing control method described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the obstacle crossing control method for a single-rod wheel-foot special vehicle described in any one of claims 1 to 7 is implemented.

Images