Wheel-legged vehicle motion trajectory planning method and device, medium and product

Abstract

The invention discloses a wheel-legged vehicle motion trail planning method and device, a medium and a product, and relates to the field of motion trail planning. The method comprises the steps that basic information of the wheel-legged vehicle in the moving process is acquired; constructing a wheel-legged vehicle model; screening landing contact points according to the wheel-legged vehicle model and the basic information by adopting a search algorithm based on a two-step method, and determining a landing contact point set; based on the landing contact point set, trajectory optimization is carried out by adopting a nonlinear constraint solution method, and an optimized trajectory is obtained; according to the invention, the motion track of the wheel-legged vehicle can be planned and controlled.

Description

Technical Field

[0001] The present application relates to the field of motion trajectory planning, and particularly to a motion trajectory planning method, device, medium and product for a wheel-leg vehicle. Background Art

[0002] Research in the field of quadruped robots has become increasingly popular recently. Many researchers have carried out research and analysis on the high-dynamic motion of such legged robots to improve the terrain passing ability that wheeled platforms lack. However, on flat terrain, the walking efficiency of quadruped robots is far lower than that of wheeled platforms. The wheel-leg vehicle combines the advantages of wheeled platforms and quadruped robots. It can avoid the defect that wheeled platforms cannot cross obstacles through legged motion. Under good terrain conditions, it can improve the moving speed through wheeled drive and has strong terrain adaptability. Through jumping motion, a large-distance displacement forward and upward, the wheel-leg vehicle can jump over relatively high obstacles or relatively far pits. Therefore, this motion has an important impact on the terrain passing ability and high-dynamic motion ability of the wheel-leg vehicle.

[0003] The control problem of jumping motion is often transformed into the problem of centroid trajectory planning, and then the control of the centroid attitude is added. At the same time, the trajectory planning and control of the wheel end are considered to ensure the attitude stability of the wheel-leg vehicle during the jumping process. When the wheel-leg vehicle encounters an obstacle that is difficult to cross by jumping, it can achieve obstacle crossing by combining jumping and climbing, that is, the wheel-leg vehicle first jumps to the edge of the obstacle, and then swings its legs and drives the wheels to climb onto the obstacle. This method can complete obstacle crossing when the joint performance of the wheel-leg vehicle is insufficient and reduce the jumping energy consumption. Therefore, how to realize the trajectory planning and control of the motion of the wheel-leg vehicle is crucial. Summary of the Invention

[0004] The purpose of the present application is to provide a motion trajectory planning method, device, medium and product for a wheel-leg vehicle, which can realize the trajectory planning and control of the motion of the wheel-leg vehicle.

[0005] To achieve the above purpose, the present application provides the following solution: A motion trajectory planning method for a wheel-leg vehicle, the method includes: obtaining the basic information of the wheel-leg vehicle during the motion process; the basic information includes: the distance from the centroid of the wheel-leg vehicle to the obstacle and the height of the obstacle.

[0006] Constructing a wheel-leg vehicle model; the wheel-leg vehicle model includes: a leg kinematic model and a single rigid body model; the leg kinematic model is a physical simplified model obtained by projecting the wheel-leg vehicle onto the sagittal plane; the single rigid body model is a floating base dynamics model simplified from the wheel-leg vehicle in a two-dimensional plane.

[0007] Adopt a search algorithm based on the two-step method, screen the landing contact points according to the wheel-leg vehicle model and the basic information, and determine the set of landing contact points; the two-step method includes: three-dimensional box constraint of the vehicle body attitude and selection of the optimal landing contact point with multiple objectives.

[0008] Based on the set of landing contact points, use the method of solving non-linear constraints to optimize the trajectory and obtain the optimized trajectory; the solution of non-linear constraints includes: the cost function and the constraint conditions corresponding to the jumping obstacle task; the optimized trajectory is used to provide an obstacle-crossing motion trajectory when the wheel-leg vehicle performs jumping and climbing.

[0009] Optionally, the mathematical expression of the single rigid body model specifically includes:

[0010]

[0011] where m CoM is the total vehicle mass of the wheel-leg vehicle; I θ is the pitch moment of inertia of the wheel-leg vehicle; g is the acceleration due to gravity, α is the terrain slope angle; T θ is the pitch moment; is the additional driving load generated at the wheel end along the tangential direction of the wheel; θ is the pitch angle of the vehicle body; is the longitudinal acceleration of the vehicle mass center in the world coordinate system; is the vertical acceleration of the vehicle mass center in the world coordinate system; is the pitch angular acceleration of the vehicle body; is the longitudinal force received at the front or rear wheel end; represents the vertical force received at the front or rear wheel end; m is the front or rear leg; m = f represents the front leg; m = b represents the rear leg.

[0012] Optionally, the three-dimensional box constraint of the vehicle body attitude is determined according to the wheel-leg vehicle model and the basic information; the three-dimensional box constraint of the vehicle body attitude specifically includes:

[0013]

[0014] where ψ is the three-dimensional box constraint of the vehicle body attitude; q CaM is the degree of freedom of the vehicle body mass center; represents the distance from the vehicle mass center of the wheel-leg vehicle to the obstacle, represents the obstacle height; W x CoM is the abscissa corresponding to the position of the mass center; W Z CoM is the ordinate corresponding to the position of the mass center; Terrain is the basic information; δ x is the increment in the dimension corresponding to the horizontal axis x; δ zis the increment in the dimension corresponding to the vertical axis z; δ θ is the increment in the dimension corresponding to the pitch angle θ of the vehicle body; θ is the pitch angle of the vehicle body; Collosion(·) is a function for judging whether the leg collides with an obstacle; are the positions of the hip joints, knee joints and wheel ends of the front and rear legs in the world coordinate system; q Joint are all joint degrees of freedom; is the minimum value of all joint degrees of freedom; is the maximum value of all joint degrees of freedom; R is the wheel radius; H is the vehicle body height; ζ is the ultimate deformation of the tire; b is the rear leg; is the real number space; is the longitudinal position of the rear wheel; is the height direction position of the front wheel.

[0015] Optionally, the selection of the multi-objective optimal landing contact point specifically includes:

[0016] Determine the objective function; the objective function includes: a stability function and an energy function; the stability function is determined based on the position of the zero moment point relative to the front and rear wheel support key points; the energy function is determined based on the energy consumption required to overcome gravity in the vertical direction during the jumping motion of the wheel-leg vehicle.

[0017] Perform sorting and screening processing according to the three-dimensional box constraint of the vehicle body posture and the objective function to obtain the vehicle body posture space set.

[0018] According to the three-dimensional box constraint of the vehicle body posture, traverse each vehicle body posture in the vehicle body posture space set and calculate the position vector; the position vector is the position vector of the wheel end relative to the centroid in the world coordinate.

[0019] According to the position vector and the total objective function, perform multi-objective optimal landing contact point selection to determine the landing contact point set; the total objective function is the sum of the stability function and the energy function.

[0020] Optionally, the cost function specifically includes:

[0021]

[0022] Among them, represents the weighted 2-norm; Q t is the weight matrix for the entire trajectory; Q l is the weight matrix for the trajectory point at the moment of landing; Q v is the weight matrix for the velocity term; is the optimized trajectory; is at the landing moment T L when the optimized trajectory; pCoM is the trajectory at the actual landing; J i is the cost function corresponding to the i-th step; N is the total number of steps; p CoM (T L ) is the actual landing trajectory at the landing moment T L ; is p CoM (T L )'s first derivative.

[0023] Optionally, the constraint conditions include: single rigid body dynamics constraints, ZMP constraints in the stretching stage, LAP point constraints, initial configuration, final configuration, joint angle limits, joint torque limits, contact sequence limits, set limits related to obstacle avoidance, and foot-end wheel positions when contacting the obstacle edge in the landing and climbing stages.

[0024] Optionally, the single rigid body dynamics constraints specifically include:

[0025]

[0026] where T θ is the pitching moment; W r m is the position vector of the wheel end relative to the center of mass in the world coordinate system; is the position vector of the ground contact point relative to the center of mass in the world coordinate system; is the additional driving load generated at the wheel end along the wheel tangent direction; α is the terrain slope angle; f m is the wheel end friction force vector of the front leg or hind leg; m is the front leg or hind leg; m = f for the front leg; m = b for the hind leg.

[0027] A computer device includes: 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 wheel-legged vehicle motion trajectory planning method.

[0028] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, it implements the above-mentioned wheel-legged vehicle motion trajectory planning method.

[0029] A computer program product includes a computer program, and when the computer program is executed by a processor, it implements the above-mentioned wheel-legged vehicle motion trajectory planning method.

[0030] According to the specific embodiments provided by the present application, the following technical effects are disclosed: The present application discloses a method, device, medium and product for planning the motion trajectory of a wheel-legged vehicle. Aiming at the obstacle-crossing method combining jumping and climbing, the obstacle information and the body pose are combined with the dynamic model of the wheel-legged vehicle. Considering two factors of stability and energy consumption economy, the optimal landing contact point (LAP point) is planned, and the optimal jumping trajectory is obtained by using non-linear optimization solution, providing a basis for the wheel-legged vehicle to combine jumping and climbing to cross obstacles. Therefore, it is possible to realize the planning and control of the motion trajectory of the wheel-legged vehicle. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] 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 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.

[0032] Figure 1 It is a schematic flowchart of the method for planning the motion trajectory of the wheel-legged vehicle provided by the embodiment of the present application.

[0033] Figure 2 It is a step flowchart of the method for planning the motion trajectory of the wheel-legged vehicle provided by the embodiment of the present application in actual application.

[0034] Figure 3 It is a schematic diagram of the simplified wheel-legged vehicle model provided by the embodiment of the present application.

[0035] Figure 4 It is a schematic diagram of a single-leg model taking the front leg as an example provided by the embodiment of the present application.

[0036] Figure 5 It is a flowchart of the two-step method provided by the embodiment of the present application.

[0037] Figure 6 It is a system structure block diagram provided by the embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0038] The following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, rather than all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present application.

[0039] The purpose of the present application is to provide a method, device, medium and product for planning the motion trajectory of a wheel-legged vehicle, aiming to realize the planning and control of the motion trajectory of the wheel-legged vehicle.

[0040] 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 with reference to the accompanying drawings and specific embodiments.

[0041] Embodiment 1: As Figure 1 shown, the wheel-leg vehicle motion trajectory planning method in this embodiment includes the following steps.

[0042] Step 100: Obtain the basic information of the wheel-leg vehicle during movement. The basic information includes: the distance from the center of mass of the wheel-leg vehicle to the obstacle and the height of the obstacle.

[0043] Step 200: Construct a wheel-leg vehicle model. The wheel-leg vehicle model includes: a leg kinematic model and a single rigid body model; the leg kinematic model is a physical simplified model obtained by projecting the wheel-leg vehicle onto the sagittal plane; the single rigid body model is a floating base dynamics model simplified from the wheel-leg vehicle in a two-dimensional plane.

[0044] Step 300: Adopt a search algorithm based on the two-step method to screen the landing contact points according to the wheel-leg vehicle model and the basic information, and determine the set of landing contact points. The two-step method includes: three-dimensional box constraint of the vehicle body attitude and selection of the multi-objective optimal landing contact point.

[0045] Step 400: Based on the set of landing contact points, use the method of solving non-linear constraints to optimize the trajectory and obtain the optimized trajectory. Solving non-linear constraints includes: the cost function and the constraint conditions corresponding to the jumping obstacle task; the optimized trajectory is used to provide an obstacle-crossing motion trajectory when the wheel-leg vehicle performs jumping and climbing.

[0046] The mathematical expression of the single rigid body model specifically includes the following formulas.

[0047]

[0048] Among them, m CoM is the total vehicle mass of the wheel-leg vehicle; I θ is the pitch moment of inertia of the wheel-leg vehicle; g is the acceleration due to gravity, α is the terrain slope angle; T θ is the pitch moment; is the additional driving load generated at the wheel end along the wheel tangent direction; θ is the pitch angle of the vehicle body; is the longitudinal acceleration of the vehicle center of mass in the world coordinate system; is the vertical acceleration of the vehicle center of mass in the world coordinate system; is the pitch angular acceleration of the vehicle body; is the longitudinal force received at the front or rear leg wheel end; Denote the vertical force received by the front or rear wheel end; m represents the front or rear leg; when m = f, it is the front leg; when m = b, it is the rear leg.

[0049] The three-dimensional box constraint of the vehicle body attitude is determined according to the wheel-leg vehicle model and basic information; the three-dimensional box constraint of the vehicle body attitude specifically includes the following formulas.

[0050]

[0051] Among them, Ψ is the three-dimensional box constraint of the vehicle body attitude; q CoM is the degree of freedom of the vehicle body's center of mass; represents the distance from the center of mass of the wheel-leg vehicle to the obstacle, represents the height of the obstacle; W x CoM is the abscissa corresponding to the position of the center of mass; W Z CoM is the ordinate corresponding to the position of the center of mass; Terrain is the basic information; δ x is the increment in the dimension corresponding to the horizontal axis x; δ z is the increment in the dimension corresponding to the vertical axis z; δ θ is the increment in the dimension corresponding to the pitch angle θ of the vehicle body; θ is the pitch angle of the vehicle body; Collision(·) is a function to judge whether the leg collides with the obstacle; is the position of the hip joint, knee joint and wheel end of the front and rear legs in the world coordinate system; q Joint is all the joint degrees of freedom; is the minimum value of all the joint degrees of freedom; is the maximum value of all the joint degrees of freedom; R is the wheel radius; H is the vehicle body height; ζ is the ultimate deformation of the tire; b is the rear leg; is the real number space; The longitudinal position of the rear wheel; is the height direction position of the front wheel.

[0052] In one embodiment, the selection of the multi-objective optimal landing contact point specifically includes the following steps.

[0053] Determine the objective function; the objective function includes: a stability function and an energy function; the stability function is determined based on the position of the zero moment point relative to the front and rear wheel support points; the energy function is determined based on the energy consumed by the wheel-leg vehicle to overcome gravity in the vertical direction during the jumping motion.

[0054] Perform sorting and screening processing according to the three-dimensional box constraint of the vehicle body attitude and the objective function to obtain the vehicle body attitude space set.

[0055] According to the three-dimensional box constraint of the vehicle body attitude, traverse each vehicle body attitude in the set of vehicle body attitude spaces, and calculate the position vector; the position vector is the position vector of the wheel end relative to the centroid in the world coordinates.

[0056] Based on the position vector and the total objective function, perform multi-objective optimal landing contact point selection to determine the set of landing contact points; the total objective function is the sum of the stability function and the energy function.

[0057] The cost function specifically includes the following formula.

[0058]

[0059] Among them, represents the weighted 2-norm; Q t is the weight matrix for the entire trajectory; Q l is the weight matrix for the trajectory point at the moment of landing; Q v is the weight matrix for the velocity term; is the optimized trajectory; is the optimized trajectory at the landing time T L ; p CoM is the actual trajectory at the time of landing; J i is the cost function corresponding to the i-th step size; N is the total number of step sizes; p CoM (T L ) is the actual landing trajectory at the landing time T L ; is p CoM (T L )'s first derivative.

[0060] The constraint conditions include: single rigid body dynamics constraints, ZMP constraints in the stretching stage, LAP point constraints, initial configuration, final configuration, joint angle limits, joint torque limits, contact sequence limits, set limits related to obstacle avoidance, and the foot end wheel positions when contacting the obstacle edge during the landing and climbing stages.

[0061] The single rigid body dynamics constraints specifically include the following formula.

[0062]

[0063] Among them, T θ is the pitching moment; W r m is the position vector of the wheel end relative to the centroid in the world coordinate system; is the position vector of the ground contact point relative to the centroid in the world coordinate system; is the additional driving load generated at the wheel end along the tangential direction of the wheel; α is the terrain slope angle; m is the front leg or hind leg; m = f represents the front leg; m = b represents the hind leg; f mThe wheel-end friction force vector of the front or rear leg.

[0064] In practical applications, for the step flowchart corresponding to the method provided in this application, see Figure 2 , as follows.

[0065] Step 1: Rely on sensors to measure the distance from the center of mass of the wheel-legged vehicle to the obstacle and the height of the obstacle.

[0066]

[0067] Among them, represents the distance from the center of mass of the wheel-legged vehicle to the obstacle, represents the height of the obstacle. Terrain is the basic information.

[0068] Step 2: Establish the forward kinematic model of the legs of the wheel-legged vehicle and the single-rigid body model, laying a model foundation for the model constraints in subsequent optimization and solution.

[0069] ① Establish the leg kinematic model. Project the wheel-legged vehicle onto the sagittal plane to simplify the model. The simplified wheel-leg vehicle model is as shown in Figure 3 .

[0070] {W} represents the world coordinate system, the floating coordinate system {B} is fixedly connected to the center of mass of the vehicle body, and θ is the pitch angle of the vehicle body. The rotation matrix W R B is as follows.

[0071]

[0072] respectively represent the ground forces received by the front and rear wheels. The vector contains the above ground forces, that is, is the vector composed of the ground forces received by the front wheel; is the vector composed of the ground forces received by the rear wheel. The degrees of freedom of the vehicle body center of mass can be expressed as Figure 4 is a single-leg model taking the front leg as an example, is the generalized coordinate vector of the hip joint, knee joint, and wheel of the front leg. When the subscript in the lower right corner is b, it correspondingly represents the joints (hip joint, knee joint) and wheel generalized vectors of the rear leg. All joint degrees of freedom can be expressed as All wheel degrees of freedom can be expressed as

[0073] In this 2D plane, the state vector q and the generalized velocity vector u of the wheel-legged vehicle are expressed as follows.

[0074]

[0075] Among them W x CoM W Z CoM T is the position of the centroid, W vx CoM W vZ CoM T describes the velocity of the centroid. is the transpose of the degrees of freedom of the vehicle body centroid; T is the transpose; is the transpose of the degrees of freedom of the joint; is the transpose of the degrees of freedom of the wheel; θ is the pitch angle of the vehicle body; is the first derivative of θ; is for q Joint 's first derivative; is for q Wheel 's first derivative.

[0076] The forward leg kinematics is used to describe the relative position relationship between the wheel end and the hip joint of the leg. Figure 4 The key symbols are marked with the front leg as an example. The structural parameters of the wheel-leg vehicle are known and unchanged, and there is no relative position difference between its hip joint and the centroid in the Z-direction floating base coordinate system. Therefore, the feasible touchdown region can be analyzed using the single-leg kinematics in the floating base coordinate system. The position vector of the knee joint relative to the hip joint and the position vector of the wheel end relative to the hip joint are:

[0077]

[0078] Taking the partial derivative of in Equation (5) gives the Jacobian matrix J m .

[0079]

[0080] Among them, is the state vector of the knee joint; l u is the distance from the hip joint to the knee joint; l l is the distance from the hip joint to the wheel end (wheel); is the state vector of the hip joint.

[0081] The positions of the hip joint, knee joint, and wheel end in the world coordinate system can more intuitively judge the collision situation between the leg and the obstacle and can be described as follows.

[0082]

[0083] In the formula, l​​h is the distance from the hip joint to the center of mass. is the position of the knee joint; is the position of the hip joint; is the position of the wheel end (wheel).

[0084] ② Establish a single-rigid-body model of the wheel-legged vehicle. The floating-base dynamics model simplified by the two-dimensional plane still has a certain degree of nonlinearity. Applying formula (5) to a large number of iterative optimization problems will increase the computational workload. Therefore, a dynamics model as linear as possible needs to be established. The single-rigid-body model has been proven feasible in the application of quadruped robots. Within the extent of not losing the dynamic relationship, this model ignores the mass of the leg system and only considers the force on the vehicle body.

[0085] The wheels of the wheel-legged vehicle can be actively driven, so an additional driving load along the tangent direction of the wheel will be generated at the wheel end. The contact model between the wheel and the obstacle adopts the rigid-body contact form. This model will only have a large difference from the wheel-leg model under the extreme yaw motion condition. The problems involved in this paper are carried out in the sagittal plane and do not need to pay attention to the lateral and yaw motions. Therefore, the rigid-body-to-rigid-body contact form can meet the requirements. After ignoring the small moment of inertia of the wheel, this additional force is expressed as follows under the ideal wheel contact state.

[0086]

[0087] In the formula, R is the wheel radius. is the rated torque of the wheel.

[0088] Then, based on the standard single-rigid-body dynamics equation, changes are made, and this model is defined as follows.

[0089]

[0090] In the formula, m CoM , I θ are respectively the total vehicle mass and the pitch moment of inertia of the wheel-legged vehicle, g is the acceleration due to gravity, and α is the terrain slope angle. In the problems studied in this application there are only two values. The pitch moment T θ is given by formula (9).

[0091]

[0092] In the formula, represents the position vector of the wheel end relative to the center of mass in the world coordinate system, represents the position vector of the ground contact point relative to the center of mass in the world coordinate system, It is the cross product operation of two-dimensional spatial vectors. It should be emphasized that when the contact sequence of each leg is 0, it means that the leg is in the air, and at this time f m = 0 and at the same time

[0093] Step 3: Find the optimal landing contact point (LAP point) based on the two-step method. Determining the position, attitude, and wheel end position of the center of mass at the moment of landing is very crucial. For the convenience of subsequent description, the point at the moment of landing is defined as the landing attachment point (LAP). The physical quantities included in this point are expressed as Equation (10).

[0094]

[0095] Among them, W x land , W z land represent the sagittal plane position of the landing point in the world coordinate system, W θ land is the pitch angle of the wheel-legged vehicle at the landing point in the world coordinate system, represents the position vectors of the wheel ends of the front leg and the rear leg relative to the center of mass of the wheel-legged vehicle in the world coordinate system.

[0096] As a key transition point, the LAP needs to consider the energy consumption factor and stability factor of the wheel-legged vehicle. Based on the theory of the stable region of the support point polygon, the position of the LAP point needs to meet at least the following conditions.

[0097]

[0098] Among them, H is the vehicle body height. Note that this is not the standing height here. If the static stability of the wheel-legged vehicle at the moment of landing is to be ensured, then the values describing the vehicle body attitude of the LAP point need to be reduced as much as possible because these physical quantities are directly related to the potential energy and kinetic energy of the wheel-legged vehicle. To find a suitable LAP and reduce the search time, this application constructs a search algorithm of the two-step method. The flow chart of the two-step method is as Figure 5 shown.

[0099] ①Construct a three-dimensional box constraint for the vehicle body attitude. By setting kinematic-related constraints, the state set meeting the conditions is collected. This state set can be expressed as follows.

[0100]

[0101] In the formula, δ x , δ z , δ θare the increments in each dimension, and their inequalities constitute the box constraints. Collision(·) is a function to determine whether the leg collides with an obstacle. represent the positions of the hip joints, knee joints, and wheel ends of the front and rear legs in the world coordinate system. The basic principle of the collision function is that in the sagittal plane, the obstacle edge can be regarded as two line segments intersecting at the vertex of the obstacle, and the collision judgment can be transformed into the problem of whether the line segments formed by the hip and knee joints and the line segments formed by the knee joints and the wheel ends intersect with the obstacle edge line segments. In addition, the front and rear wheels are required to be close to the obstacle edge, and the wheels cannot be squeezed beyond the limit deformation ζ of the tire. At the same time, the range constraints of the joint angles are necessary. In the process of constructing the state set containing the above constraints, although the traversal amount of the posture and joint angles is large, the calculations involved here are simple.

[0102] ② Selection of the LAP point based on multi-objective optimization. After constructing the qualified ψ set, two objectives of optimal stability and minimum energy consumption are selected to perform more detailed screening on the elements in the set. Compared with the centroid projection point, the zero moment point can more accurately characterize the stability of the wheel-leg vehicle. When the wheel-leg vehicle lands at the LAP point, the pitching inertia of the vehicle body will cause the instability of the vehicle state, and the zero moment point (ZeroMomentpoint, ZMP) is very likely to be outside the line connecting the front and rear support points. The actuation of the wheels will play a key role at this stage, and the change in the position of the ZMP point caused by the additional force provided by the wheels will be considered. In this problem, only the position of the ZMP point in the sagittal plane at the landing moment is concerned. Assuming that the wheel ends of the front and rear legs are in good contact with the obstacle edge at this moment, according to the definition of the ZMP, the relative position in the X direction between this point and the centroid of the wheel-leg vehicle in the world coordinate system W x ZMP is as follows.

[0103]

[0104] When the ZMP point is on the line connecting the support points, it represents that the state of the wheel-leg vehicle is relatively stable at this time. The ZMP of static analysis can reflect the sensitivity of the measure of using wheel actuation to improve vehicle stability to different centroid postures and leg postures, and indirectly reflect the level of stability. Therefore, it is defined that the stability score depends on the position of the ZMP point relative to the midpoint of the front and rear wheel supports. A higher score for this item indicates that using wheel actuation in the current posture can better avoid the vehicle from tipping backward and losing stability. Stability score Score ZMP is expressed as follows.

[0105]

[0106] where Wr f (1) is the position of the rear wheel end relative to the centroid in the world coordinate system.

[0107] Based on comprehensive considerations, an energy minimization score is also proposed. It mainly depends on the relative position of the center of mass height of the wheel-legged vehicle within the box constraints because the vehicle needs to overcome gravity in the vertical direction during the jumping motion, and its energy consumption accounts for the vast majority compared to the other two degrees of freedom of the center of mass in the sagittal plane. The energy minimization score Score Energy can be expressed as follows.

[0108]

[0109] The total score Score Sum is expressed as the sum of two objective scores and can be expressed as follows.

[0110] Score Sum = λ ZMP Score ZMP + λ Energy Score Energy (16).

[0111] Among them, λ ZMP , λ Energy are respectively the weights of the stability score and the energy minimization score.

[0112] First, set the ground stiffness γ, the friction coefficient μ, and the rated torque of the wheel motor Then, sort the elements in the ψ space in ascending order according to W z CoM and select k groups of smaller body postures (degrees of freedom of the center of mass of the vehicle) q CoM to form a new space Ω. Traverse each group of postures in the new space to calculate the W r m that meets the constraints, and calculate the comprehensive score (total score) Score Sum , and select the group of data with the highest score as the LAP point.

[0113] Step 4: Optimization of the jumping trajectory based on the non-linear problem. For a wheel-legged vehicle, the key to achieving a jump is appropriate trajectory planning. An optimization problem is established to find the trajectory of the attitude position of the center of mass, including the horizontal and vertical positions and the roll angle of the vehicle body, and at the same time, the trajectory of the wheel end position also needs to be planned. During the entire process of a wheel-legged vehicle jumping over an obstacle, it is difficult to have a suitable trajectory in advance for optimization, but it is feasible to set the trajectory points at important moments during the motion process. Therefore, set the expected states of the wheel-legged vehicle at the initial moment T J , the landing moment T L and the end moment T S of the stretching process. That is is the desired state of the wheel-legged vehicle at the initial moment T of the jumping process J ; is the desired state of the wheel-legged vehicle at the landing moment T L ; is the desired state of the wheel-legged vehicle at the end moment T of the extension process. Without loss of generality, the trajectory used for optimization S can be expressed as follows.

[0114]

[0115] where, PLInterp(·) is a piecewise linear interpolation operation.

[0116] The cost function adopts the conventional quadratic form, and this nonlinear optimization problem can be defined as in Equation (18). Specifically, it includes the following three parts. First is to minimize:

[0117]

[0118] Then, find:

[0119]

[0120] Finally, make it subject to:

[0121] c(X) ≤ 0 h(X) = 0.

[0122] where, represents the weighted 2-norm, and Q t , Q l are the weight matrix of the entire trajectory and the weight matrix of the trajectory points at the landing moment respectively. is the transpose of the position vector of the front wheel end relative to the center of mass in the world coordinate system; is the transpose of the position vector of the rear wheel end relative to the center of mass in the world coordinate system.

[0123] Since the subsequent extension process needs to be carried out based on the state that the wheel-legged vehicle reaches the LAP point stably and smoothly, the weight at the landing moment needs to be set to a relatively large value so that the obtained trajectory fits the state set by LAP as closely as possible at the moment T L . A quadratic term of the landing moment velocity is introduced in the cost function to enable the vehicle to land with less kinetic energy, and Q v is the weight matrix of the velocity term. The optimization variable X i of the i-th step is associated with the single-rigid-body dynamics. c(X) and h(X) represent inequality constraints and equality constraints respectively. The optimization step is expressed as step = T S / N.

[0124] ​For the sake of convenience, the origin of the world coordinate system is set at the ground projection point of the center of mass of the wheel-leg vehicle at the optimized initial moment. Various constraints required for the jump obstacle task can be described as follows.

[0125] 1. Single-rigid-body dynamics constraint, see Equation (9).

[0126] 2. ZMP constraint in the extension stage.

[0127] 3. LAP point constraint: Where

[0128] 4. Initial configuration:

[0129] q CoM (TJ) is the degree of freedom of the center of mass of the vehicle body at the initial moment T J during the jump process; is the vertical coordinate of the position of the standard center of mass.

[0130] 5. Final configuration:

[0131] q CoM (T S ) is the degree of freedom of the center of mass of the vehicle body at the end moment T S during the extension process.

[0132] 6. Joint angle limit:

[0133] 7. Joint torque limit: τ max is the maximum joint torque.

[0134] 8. Contact sequence limit: The contact sequence Cont = 1 during the preparation stage, landing stage and extension stage 1×4 ; the contact sequence Cont = 0 during the jump stage 1×4 .

[0135] 9. Set limit related to obstacle avoidance.

[0136] 10. The position of the wheel at the foot end when contacting the edge of the obstacle during the landing and climbing stages.

[0137] More precisely, the optimization problem needs to satisfy the dynamic model, which will make the planned trajectory within the motion and driving capabilities of the wheel-leg vehicle. Here, the single-rigid-body dynamic equation (9) is used as a constraint. The ZMP constraint in the extension stage is set, and the ZMP in the X direction is obtained through equation (15) and is expected to be located between the two wheel ends to obtain a trajectory that stabilizes the vehicle body. The LAP constraint will take effect before the landing moment. It is hoped that the wheel-leg vehicle can reach the expected LAP point 5 steps in advance and maintain this position in the next few iterations. The initial configuration constraint is expressed as the attitude of the wheel-leg vehicle in the normal state, which is the height of the vehicle's center of mass in the world coordinate system in the normal state. The terminal configuration constraint is also considered. At the end of the extension stage of the wheel-leg vehicle, the pitch angle and the standing height of the vehicle body are set to be the same as those at the initial moment. A soft constraint is set in the horizontal direction, which is very important. The horizontal distance traveled by the vehicle during the entire extension stage is related to the wheel speed, and the magnitude of the wheel speed cannot be strictly constrained at this stage because the wheel speed is associated with the wheel torque. Instead, more emphasis is placed on the wheel torque that keeps the vehicle stable through the ZMP constraint. There is a kinematic relationship between the joint angle constraint and W r m . The torque limit constraint is associated with equations (6) and (9), and the torque can be expressed by the ground reaction force solved by the single-rigid-body model and the Jacobian matrix. The contact sequence Cont specifies the ground contact situation of the wheel ends in each stage. The geometric constraint is used to avoid collisions between the legs and the body of the wheel-leg vehicle with obstacles. Here, the function Collision(·) is also used. The wheel-end position constraints in the landing stage and the extension stage require the wheels to be close to the edge of the obstacle. The above nonlinear programming problem is established with the help of the nonlinear optimization tool CasADi, and then the multiple shooting method algorithm is written to solve it.

[0138] This application also provides a wheel-leg vehicle motion trajectory planning system, which includes the following modules.

[0139] The obstacle information extraction module analyzes the obstacle information and extracts information such as the obstacle height and the horizontal distance from the vehicle's center of mass to the obstacle, providing a basis for subsequent trajectory planning information.

[0140] The landing contact point selection module constructs a three-dimensional box constraint for the vehicle body attitude in combination with the wheel-leg vehicle dynamic model and finds the optimal landing contact point based on the two-step method.

[0141] The jump trajectory optimization module sets important trajectory points in advance during the jump, and in combination with the vehicle dynamic model constraints, establishes an optimization problem to find the attitude position trajectory of the center of mass and the wheel-end position trajectory. The system structure block diagram is as Figure 6 shown.

[0142] In order to reduce the ability requirements of the joint torque during the jumping process and improve the ability of the wheel-legged vehicle to jump over obstacles, the present application provides a motion trajectory planning method for the wheel-legged vehicle to jump over obstacles by integrating climbing actions. First, a new jumping method and the floating base dynamics and kinematics models of the wheel-legged vehicle are described in detail, and a single-rigid-body model considering the wheel driving force is established for model constraints in nonlinear problems. Secondly, an optimal landing point state searcher based on a two-step method and a trajectory optimizer based on a nonlinear programming method are proposed. The two constitute a motion planner, which can offline generate the centroid trajectory and wheel-end trajectory that meet the dynamic characteristics of the wheel-legged vehicle, the actuator capabilities, the ZMP constraint, and the terrain constraint, etc., for jumping.

[0143] Quadruped wheel-legged vehicles have both the terrain adaptability of quadruped platforms and the high-speed movement ability of wheeled platforms, and are widely used. Usually, when a wheel-legged vehicle jumps over an obstacle, it will choose to directly jump over the obstacle. However, when encountering a relatively large obstacle, it is difficult to directly jump over the obstacle. At this time, the combination of jumping and climbing actions can be used to achieve obstacle crossing, which can reduce the requirements for joint torque performance during the jumping process and improve the jumping obstacle-crossing ability of the wheel-legged vehicle. However, there is very little research on the jumping combined with climbing obstacle crossing of wheel-legged vehicles at present. The motion trajectory planning method for the wheel-legged vehicle to jump over obstacles by integrating climbing actions proposed in the present application can select the optimal landing contact point by obtaining the obstacle size information and the wheel-legged vehicle's own position information, and use nonlinear optimization to obtain the optimal trajectory, which can be used for the jumping obstacle crossing of the wheel-legged vehicle by integrating climbing.

[0144] The advantages corresponding to the method proposed in the present application are as follows.

[0145] The obstacle crossing of the wheel-legged vehicle by combining jumping and climbing actions enables it to cross more difficult obstacles, effectively improving the terrain adaptability of the vehicle.

[0146] Integrating the dynamic model of the wheel-legged vehicle into the physical constraints of the trajectory planning ensures that the planned trajectory complies with the basic dynamic performance of the wheel-legged vehicle. While fully considering the body attitude stability of the wheel-legged vehicle during the jumping process, the energy consumption is also considered in the trajectory planning, so that the energy consumption during the jumping combined with climbing process is minimized as much as possible, improving the energy consumption economy of the vehicle as much as possible while ensuring the safety of the vehicle.

[0147] Embodiment 2

[0148] A computer device includes: a memory, a processor, and a computer program stored on the memory and executable on the processor. The processor executes the computer program to implement the wheel-legged vehicle motion trajectory planning method in Embodiment 1.

[0149] Embodiment 3

[0150] A computer-readable storage medium stores a computer program thereon. When the computer program is executed by a processor, it implements the wheel-legged vehicle motion trajectory planning method in Embodiment 1.

[0151] Embodiment 4

[0152] A computer program product includes a computer program. When the computer program is executed by a processor, it implements the wheel-legged vehicle motion trajectory planning method in Embodiment 1.

[0153] Embodiment 5

[0154] A computer device, which can be a database. 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 transactions to be processed. 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 the wheel-legged vehicle motion trajectory planning method in Embodiment 1.

[0155] It should be noted that the object information (including but not limited to object device information, object 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 object or fully authorized by all parties, and the collection, use, and processing of relevant data need to comply with the relevant laws, regulations, and standards of relevant countries and regions.

[0156] Those of ordinary skill in the art can understand that all or part of the processes in the methods of 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 can include the processes of the embodiments of the above methods. Among them, any reference to a memory, database, or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memories. Non-volatile memory can 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 can 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. The databases involved in the embodiments provided in the present application can include at least one of relational databases and non-relational databases. Non-relational databases can include distributed databases based on blockchain, etc., without limitation. The processors involved in the embodiments provided in the present application can be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, data processing logics based on quantum computing, etc., without limitation.

[0157] 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 recorded in this specification.

[0158] Specific examples are used in this article to elaborate on the principles and implementation manners of the present application. The descriptions of the above embodiments are 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 method for planning the motion trajectory of a wheel-legged vehicle, characterized in that: The method comprises: Obtaining basic information of the wheel-legged vehicle during movement; the basic information includes: the distance from the center of mass of the wheel-legged vehicle to the obstacle and the height of the obstacle; Constructing a wheel-legged vehicle model; the wheel-legged vehicle model includes: a leg kinematics model and a single rigid body model; the leg kinematics model is a physically simplified model obtained by projecting the wheel-legged vehicle onto the sagittal plane; the single rigid body model is a floating basis dynamics model based on a two-dimensional plane simplification of the wheel-legged vehicle; A search algorithm based on a two-step method is used to screen the landing contact points according to the wheel-leg vehicle model and the basic information to determine the landing contact point set; the two-step method includes: vehicle body posture three-dimensional box constraint and multi-objective optimal landing contact point selection; Based on the landing contact point set, a nonlinear constraint solving method is used to perform trajectory optimization to obtain an optimized trajectory; the nonlinear constraint solving includes: a cost function and constraint conditions corresponding to the obstacle jumping task; the optimized trajectory is used to provide a motion trajectory for overcoming obstacles when the wheel-legged vehicle jumps and climbs.

2. The method for planning the motion trajectory of a wheel-legged vehicle according to claim 1, characterized in that: The mathematical expression of the single rigid body model specifically includes: Among them, m CoM I is the vehicle mass of the wheel-legged vehicle; θ is the pitch moment of inertia of the wheel-legged vehicle; g is the acceleration of gravity, α is the terrain slope angle; T θ is the pitching moment; Generates additional driving load along the wheel tangent direction for the wheel end; θ is the pitch angle of the vehicle body; is the longitudinal acceleration of the vehicle's center of mass in the world coordinate system; is the vertical acceleration of the vehicle's center of mass in the world coordinate system; is the pitch angular acceleration of the vehicle body; It is the longitudinal force on the front or rear wheel end; It indicates the vertical force on the front or rear wheel end; m is the front or rear leg; when m=f, it is the front leg; when m=b, it is the rear leg.

3. The method for planning the motion trajectory of a wheel-legged vehicle according to claim 1, characterized in that: The body posture three-dimensional box constraint is determined according to the wheel-leg vehicle model and the basic information; the body posture three-dimensional box constraint specifically includes: Among them, ψ is the three-dimensional box constraint of the vehicle body posture; q CaM is the degree of freedom of the center of mass of the vehicle body; represents the distance from the center of mass of the wheel-legged vehicle to the obstacle, Indicates the obstacle height; is the horizontal coordinate corresponding to the position of the center of mass; is the vertical coordinate corresponding to the position of the center of mass; Terrain is the basic information; δ x is the increment on the dimension corresponding to the horizontal axis x; δ z is the increment of the dimension corresponding to the vertical axis z; δ θ is the increment of the pitch angle θ of the vehicle body in the corresponding dimension; θ is the pitch angle of the vehicle body; Collision(·) is the function for determining whether the leg collides with the obstacle; is the position of the hip joint, knee joint and wheel end of the front and rear legs in the world coordinate system; q Joint for all joint degrees of freedom; is the minimum value of all joint degrees of freedom; is the maximum value of all joint degrees of freedom; R is the wheel radius; H is the height of the vehicle body; ζ is the limit deformation of the tire; b is the rear leg; is the real number space; is the longitudinal position of the rear wheel; It is the height direction position of the front wheel.

4. The method for planning the motion trajectory of a wheel-legged vehicle according to claim 3, characterized in that: The multi-objective optimal landing contact point selection specifically includes: Determine an objective function; the objective function includes: a stability function and an energy function; the stability function is determined based on the position of the zero moment point relative to the front and rear wheel support points; the energy function is determined based on the energy consumption of the wheel-legged vehicle in the vertical direction to overcome gravity during the jumping movement; Sorting and screening are performed according to the vehicle body posture three-dimensional box constraint and the objective function to obtain a vehicle body posture space set; According to the three-dimensional box constraint of the vehicle body posture, traverse each vehicle body posture in the vehicle body posture space set to calculate a position vector; the position vector is a position vector of the wheel end relative to the center of mass in the world coordinate; According to the position vector and the sum of the objective functions, multi-objective optimal landing contact point selection is performed to determine the landing contact point set; the sum of the objective functions is the sum of the stability function and the energy function.

5. The method for planning the motion trajectory of a wheel-legged vehicle according to claim 1, characterized in that: The cost function specifically includes: in, represents the weighted 2-norm; Q t is the weight matrix of the whole trajectory; Q l is the weight matrix of the trajectory points at the moment of landing; Q v is the weight matrix of the velocity term; To optimize the trajectory; is the time of landing T L The optimal trajectory when p CoM is the actual trajectory when landing; J i is the cost function corresponding to the i-th step length; N is the total number of steps; p CoM (T L ) is the time of landing T L The actual landing trajectory at ; For p CoM (T L ) is the first-order derivative of .

6. The method for planning the motion trajectory of a wheel-legged vehicle according to claim 1, characterized in that: The constraints include: single rigid body dynamics constraints, ZMP constraints in the extension phase, LAP point constraints, initial configuration, final configuration, joint angle constraints, joint torque constraints, contact sequence constraints, set constraints related to obstacle avoidance, and foot-end wheel positions when contacting the edge of obstacles during landing and climbing phases.

7. The method for planning the motion trajectory of a wheel-legged vehicle according to claim 6, characterized in that: The single rigid body dynamics constraints specifically include: Among them, T θ is the pitching moment; is the position vector of the wheel end relative to the center of mass in the world coordinate system; is the position vector of the ground contact point relative to the center of mass in the world coordinate system; The wheel end generates an additional driving load along the wheel tangent direction; α is the terrain slope angle; f m is the wheel end friction force vector of the front leg or the rear leg; m is the front leg or the rear leg; when m=f, it is the front leg; when m=b, it is the rear leg.

8. 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 method for planning the motion trajectory of a wheel-legged vehicle as described in any one of claims 1-7.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for planning the motion trajectory of a wheel-legged vehicle described in any one of claims 1 to 7 is implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the method for planning the motion trajectory of a wheel-legged vehicle described in any one of claims 1 to 7 is implemented.

Images