A quadruped robot adaptive jumping control method

Abstract

This invention provides an adaptive jumping control method for a quadruped robot. The method includes: acquiring the elevation information of a target obstacle and calculating the target pose of the robot trajectory based on the elevation information; solving a pre-constructed target optimization function according to set constraints to obtain the target motion trajectory of the robot jumping over the target obstacle; generating a safe landing position for the swing leg based on the current position and the target position of the swing leg; calculating the swing trajectory of the swing leg based on the safe landing position and the target motion trajectory; constructing a trajectory tracking model using the target motion trajectory as input, and solving the trajectory tracking model with a minimum cost function to obtain the optimized foot force of the supporting leg; and calculating the joint motion parameters under the target pose of the robot trajectory using the joint compliant servo control model. This method solves the problems of poor stability and high risk of tipping over when the robot crosses large obstacles.

Description

Technical Field

[0001] This invention relates to the field of robot control technology, and in particular to an adaptive jumping control method for a quadruped robot. Background Technology

[0002] Compared to traditional wheeled and tracked vehicles, quadruped robots can adapt to complex terrains such as plateaus and mountains by selecting discrete landing points for their swinging legs. Currently, quadruped robots mainly rely on walking to traverse or avoid obstacles. When encountering large obstacles, walking to traverse obstacles has a limited range of motion and poor stability after landing, while avoiding obstacles increases the path length, resulting in low task execution efficiency. Therefore, in existing technologies, jumping can be used to enable robots to overcome obstacles. However, when robots use jumping to overcome obstacles, it is usually done remotely. During the control operation, the relative position information of the obstacle cannot be determined, making the robot prone to tipping over.

[0003] In view of this, providing an adaptive jumping control method for quadruped robots to solve the problems of poor stability and high risk of tipping over when robots cross large obstacles, thereby improving the landing stability of legged robots when crossing obstacles, reducing the risk of tipping over, and thus improving the passability of legged robots in complex terrain, has become an urgent problem to be solved by those skilled in the art. Summary of the Invention

[0004] This invention provides an adaptive jumping control method for quadruped robots to solve the problems of poor stability and high risk of tipping over when robots cross large obstacles, thereby improving the landing stability of legged robots when crossing obstacles, reducing the risk of tipping over, and thus improving the performance of legged robots in complex terrain.

[0005] This invention provides an adaptive jumping control method for a quadruped robot, comprising:

[0006] Obtain the elevation information of the target obstacle, and calculate the target pose of the computer body trajectory based on the elevation information;

[0007] Based on the set constraints, the pre-constructed target optimization function is solved to obtain the target motion trajectory of the body jumping over the target obstacle;

[0008] Based on the current position and target position of the swing leg, generate a safe landing position for the swing leg;

[0009] Calculate the swing trajectory of the swing leg based on the safe landing position and the target movement trajectory;

[0010] A trajectory tracking model is constructed using the target motion trajectory as input, and the trajectory tracking model is solved using the minimum cost function to obtain the optimized supporting leg foot end force;

[0011] Using the optimized supporting leg foot force as input, a joint compliant servo control model is constructed, and the joint motion parameters under the target pose of the computer body trajectory are calculated through the joint compliant servo control model, so that the control device of the legged robot outputs control commands according to the joint motion parameters. The control commands are used to control the leg joint movements of the legged robot.

[0012] The adaptive jumping control method for a quadruped robot provided by the present invention acquires the elevation information of a target obstacle and calculates the target pose of the robot's trajectory based on the elevation information, specifically including:

[0013] Obtain the height information of the target obstacle and the distance information of the target obstacle relative to the origin;

[0014] The position of the target aircraft is calculated based on the height information, the distance information, and the parameter information of the target aircraft.

[0015] The trajectory and pose of the target body are calculated based on the position, attitude, average velocity, and angular velocity of the target body.

[0016] According to the adaptive jumping control method for quadruped robots provided by the present invention, the parameter information of the target body includes the safe distance of the center of mass of the target body relative to the edge of the target obstacle, the length of the target body, and the width of the target body;

[0017] Using the first expression, the position of the target aircraft is calculated based on the height information, the distance information, and the parameter information of the target aircraft.

[0018] The first expression is:

[0019] p e =[d x +d m +l / 2,d y +d m +w / 2,h o ] T

[0020] Where, p e d represents the location of the target machine. x The distance information is represented by its x-axis component, d y d represents the y-component of the distance information. mThe distance between the center of mass of the target body and the edge of the target obstacle is denoted as l, where l is the length of the target body and w is the width of the target body.

[0021] According to the adaptive jumping control method for quadruped robots provided by the present invention, the pre-constructed objective optimization function is:

[0022]

[0023] Among them, Q x and R u This is the weight matrix. For the desired dynamic state of the quadruped robot system, x k This represents the state of the quadruped robot system in its current pose. For the desired control input, u k N represents the current control input, and N represents the number of iterations.

[0024] According to the adaptive jumping control method for quadruped robots provided by the present invention, the set constraints include environmental constraints and dynamic constraints.

[0025] The constraints include at least the system dynamics constraints, force constraints, friction cone constraints, foot position constraints, leg length constraints, acceleration constraints, torque constraints, and joint angle constraints of the quadruped robot.

[0026] The adaptive jumping control method for quadruped robots provided by this invention generates a safe landing position for the swing leg based on the current position and the target position of the swing leg, specifically including:

[0027] Get the current position r of the swing leg c and the target position r of the swing leg f ;

[0028] Using the second expression, based on the current position r c With the target position r f Perform linear interpolation to obtain a safe landing position;

[0029] The second expression is:

[0030] r = r c +(r f -r c ) / T*t

[0031] Where t is the oscillation time, T is the oscillation period, and t∈(0,T).

[0032] According to the adaptive jumping control method for quadruped robots provided by the present invention, the swing trajectory of the swing leg is calculated based on the safe landing position and the target motion trajectory, specifically including:

[0033] Before the aircraft lands, the starting position r1 and starting velocity of the swing leg are determined based on the safe landing position and the target motion trajectory. End point position r4, end point velocity A fifth-order polynomial is constructed using the position r2 coinciding with the edge corner of the target obstacle and the highest swing point r3;

[0034] The oscillating trajectory is obtained by solving a fifth-degree polynomial.

[0035] According to the adaptive jumping control method for quadruped robots provided by the present invention, the minimum cost function min f′(x) is:

[0036]

[0037] Among them, Q′ x and R′ u Weight matrix; For the trajectory of the target machine; u′ k The desired input is x′; T is the oscillation period; N is the number of iterations; k This represents the current trajectory of the machine.

[0038] According to the adaptive jumping control method for quadruped robots provided by the present invention, the constraints of the minimum cost function min f′(x) include at least the system dynamics constraints, force constraints, and friction cone constraints of the quadruped robot.

[0039] According to the adaptive jumping control method for quadruped robots provided by the present invention, the joint compliant servo control model is as follows:

[0040]

[0041] In the formula: K pθ K is the controller stiffness coefficient matrix; vθ θ is the controller damping coefficient matrix; θ is the actual joint angle vector. The desired joint angular velocity vector; τ is the actual joint angular velocity vector. ff =-J T u′ k is the joint force feedforward term; J is the joint force Jacobian matrix; u is the joint controller input parameter.

[0042] The adaptive jumping control method for quadruped robots provided by this invention acquires the elevation information of a target obstacle and calculates the target pose of the robot trajectory based on the elevation information; solves a pre-constructed target optimization function according to set constraints to obtain the target motion trajectory of the robot jumping over the target obstacle; generates a safe landing position for the swing leg based on the current position and the target position of the swing leg; calculates the swing trajectory of the swing leg based on the safe landing position and the target motion trajectory; constructs a trajectory tracking model using the target motion trajectory as input, solves the trajectory tracking model using a minimum cost function to obtain the optimized foot force of the supporting leg; constructs a joint compliant servo control model using the optimized foot force of the supporting leg as input, and calculates the joint motion parameters under the target pose of the robot trajectory through the joint compliant servo control model, so that the control device of the legged robot outputs control commands according to the joint motion parameters, and the control commands are used to control the leg joint movements of the legged robot.

[0043] This adaptive jumping control method for quadruped robots fully considers obstacle information and optimizes the generation of the robot's motion trajectory, thereby effectively enabling the robot to adapt to jumping. Simultaneously, by combining safe and unsafe zones within the robot's flight space for online planning of the swing leg's landing and swing trajectory, obstacles can be effectively avoided. This allows for autonomous and stable obstacle crossing even with larger obstacles, offering high safety and high passage efficiency. By fusing obstacle information, this method controls the robot to jump autonomously, achieving precise dynamic obstacle crossing and significantly improving the robot's passage efficiency and stability with large obstacles. It solves the problems of poor stability and high tipping risk when robots cross large obstacles, thereby improving the landing stability of legged robots when crossing obstacles, reducing tipping risk, and ultimately improving the performance of legged robots in complex terrain. Attached Figure Description

[0044] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0045] Figure 1 This is one of the flowcharts for the adaptive jumping control method for a quadruped robot provided by the present invention;

[0046] Figure 2 The second flowchart of the adaptive jumping control method for a quadruped robot provided by the present invention;

[0047] Figure 3This is a schematic diagram of the body's jumping trajectory in a specific application scenario of the present invention;

[0048] Figure 4 This is a schematic diagram illustrating the safe landing of the machine during a jump in a specific application scenario of the present invention;

[0049] Figure 5 This is a schematic diagram illustrating a safe landing of the aircraft during a jump in a specific application scenario of the present invention;

[0050] Figure 6 This is a schematic diagram of the body support force in a specific application scenario of the present invention;

[0051] Figure 7 This is a schematic diagram of the machine jumping over obstacles in a specific application scenario of the present invention. Detailed Implementation

[0052] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0053] The following is combined with Figures 1-7 The provided method describes an adaptive jumping control method for a quadruped robot.

[0054] In one specific implementation, such as Figure 1 As shown, the present invention provides an adaptive jumping control method for a quadruped robot, comprising the following steps:

[0055] S110: Obtain the elevation information of the target obstacle, and calculate the target pose of the computer body trajectory based on the elevation information;

[0056] S120: Based on the set constraints, solve the pre-built target optimization function to obtain the target motion trajectory of the body jumping over the target obstacle, that is, generate the jumping body trajectory that integrates terrain information;

[0057] S130: Based on the current position and target position of the swing leg, generate a safe landing position for the swing leg;

[0058] S140: Calculate the swing trajectory of the swing leg based on the safe landing position and the target movement trajectory, so as to plan the swing leg behavior online;

[0059] S150: Construct a trajectory tracking model using the target motion trajectory as input, solve the trajectory tracking model with the minimum cost function to obtain the optimized support leg foot end force, and establish a tracking control model for the swing leg and support leg;

[0060] S160: Using the optimized supporting leg foot force as input, a joint compliant servo control model is constructed, and the joint motion parameters under the target pose of the robot body trajectory are calculated through the joint compliant servo control model, so that the control device of the legged robot outputs control commands according to the joint motion parameters. The control commands are used to control the leg joint movements of the legged robot to establish an online tracking control model for the swing leg trajectory and a joint compliant servo control model, input the target trajectory of the robot body, and output the online rolling optimized supporting leg foot force.

[0061] like Figure 2 As shown, obtaining the elevation information of the target obstacle and calculating the target's trajectory and pose based on the elevation information specifically includes the following steps:

[0062] S210: Obtain the height information of the target obstacle and the distance information of the target obstacle relative to the origin; that is, use the airborne perception sensors to establish the height and distance information of the obstacle in front, and calculate the safe aircraft trajectory target pose based on the elevation information;

[0063] S220: Calculate the position of the target aircraft based on the height information, the distance information, and the parameter information of the target aircraft;

[0064] S230: Calculate the target trajectory pose of the target body based on its position, attitude, average velocity, and angular velocity; that is, establish a target optimization function that includes multiple constraints such as environmental constraints and dynamics to generate the trajectory of the body jumping over obstacles.

[0065] In S220, the parameter information of the target body includes the safe distance of the target body's center of mass relative to the edge of the target obstacle, the length of the target body, and the width of the target body;

[0066] Using the first expression, the position of the target aircraft is calculated based on the height information, the distance information, and the parameter information of the target aircraft.

[0067] The first expression is:

[0068] p e =[d x +d m +l / 2,d y +d m +w / 2,h o ]T

[0069] Where, p e d represents the location of the target machine. x The distance information is represented by its x-axis component, d y d represents the y-component of the distance information. m The distance between the center of mass of the target body and the edge of the target obstacle is denoted as l, where l is the length of the target body and w is the width of the target body.

[0070] In a specific use case, generating the target's motion trajectory, that is, generating the trajectory of a jumping machine that incorporates terrain information, such as... Figure 3 As shown, the specific steps include:

[0071] First, the height information h of the obstacle ahead is established using onboard sensing sensors. o Based on the distance d relative to the origin and the elevation information, a safe target pose for the aircraft trajectory is calculated.

[0072] Where, p e Θ represents the position of the target body calculated using the first expression above. e The orientation of the target machine. Let ω be the translational velocity of the target machine. e ω represents the angular velocity of the target machine.

[0073] Then, a target optimization function with multiple constraints including environmental constraints and dynamics is established to generate the motion trajectory of the body jumping over obstacles.

[0074] Specifically, the dynamics of the quadruped robot system are described as follows:

[0075] x k+1 =A k x k +B k f k +G k

[0076] in,

[0077]

[0078]

[0079]

[0080] In the formula, p is the position of the robot's center of mass, ω is the angular velocity of the robot's body rotation, and F... i Let r be the force exerted by the i-th supporting leg, m be the mass of the robot, and r be the force exerted by the foot. iis the location of the foot-to-ground interaction point, I is the moment of inertia of the organism, and g is the acceleration due to gravity. Let R be the Euler angular velocity, and R be the transformation matrix between the rotational angular velocity and the Euler angular velocity.

[0081] In S120 above, the pre-constructed objective function is:

[0082]

[0083] Among them, Q x and R u This is the weight matrix. For the desired dynamic state of the quadruped robot system, x k This represents the state of the quadruped robot system in its current pose. For the desired control input, u k N represents the current control input, and N represents the number of iterations.

[0084] Furthermore, the landing point and the force exerted by the foot are used as control inputs, namely:

[0085] u = [r1, F1, r2, F2, r3, F3, r4, F4]

[0086] The pre-constructed objective optimization function is a minimum cost function, and its constraints include environmental constraints and dynamic constraints; wherein, the constraints include at least the system dynamic constraints, force constraints, friction cone constraints, foot position constraints, leg length constraints, acceleration constraints, torque constraints and joint angle constraints of the quadruped robot.

[0087] Specifically, the constraints are established as follows:

[0088] stx k+1 =A k x k +B k f k +G k System Dynamics

[0089] Force Constraint

[0090] Friction cone constraint

[0091] feet on the ground

[0092] The height of the foot is lower than the height of the obstacle.

[0093] ||(r l,k -p) b -p hip ||≤l maxLeg length constraint

[0094] Height constraints

[0095] Kinematic constraints

[0096] Acceleration constraints

[0097] Torque Constraint

[0098] q min ≤|q l,j |≤q max Joint angle constraints

[0099] in, The ground normal force in the k-th iteration; p is the tangential force; μ is the coefficient of sliding friction. hip The position of the hip joint relative to the robot's center of mass; (r l,k -p) b The position of the foot relative to the center of mass of the fuselage; r lx,k r ly,k For foot position r l,k x and y direction components; d1 is the safety distance; H g_k To correspond to the terrain height for the k-th iteration, when the robot jumps, H... g_k =0, when the robot lands, H g_k =H o .

[0100] In S130 above, based on the current position and the target position of the swing leg, a safe landing position for the swing leg is generated, specifically including:

[0101] Get the current position r of the swing leg c and the target position r of the swing leg f ;

[0102] Using the second expression, based on the current position r c With the target position r f Perform linear interpolation to obtain a safe landing position; that is, plan the robot's swing landing position within a safe flight space area and perform linear interpolation by combining the current position with the target position.

[0103] The second expression is:

[0104] r = r c +(r f -r c ) / T*t

[0105] Where t is the oscillation time, T is the oscillation period, and t∈(0,T).

[0106] In S140 above, the swing trajectory of the swing leg is calculated based on the safe landing position and the target movement trajectory, specifically including:

[0107] Before the aircraft lands, the starting position r1 and starting velocity of the swing leg are determined based on the safe landing position and the target motion trajectory. End point position r4, end point velocity A fifth-order polynomial is constructed using the position r2 coinciding with the edge corner of the target obstacle and the highest swing point r3;

[0108] By solving the fifth-order polynomial, the swing trajectory is obtained. In this way, the safe landing position and swing trajectory of the robot can be planned in the area close to the obstacle, so that the robot can avoid the obstacle and land safely.

[0109] Based on the above steps S130 and S140, in the specific application scenario described above, when planning the swing leg behavior online, such as Figure 4 and Figure 5 As shown, the specific steps include:

[0110] First, the robot's swing landing position is planned within a safe flight space area. Based on the second expression mentioned above, and combined with the current position r... c With target position r f Perform linear interpolation.

[0111] Then, the robot's safe landing position and swing trajectory are planned in the area near the obstacle, so that the robot can avoid the obstacle and land safely.

[0112] Specifically, before landing, when the robot approaches the obstacle area, the original landing point r0 may collide with the obstacle. At this time, a safe landing point r1 is selected.

[0113] In this embodiment, it is assumed that the obstacle is directly in front of the robot, and at this time r1 = r0 - [d1,0,0].

[0114] As the robot approaches landing, its original landing method might result in a collision with obstacles, and its swing trajectory might also interfere with the obstacles. A fifth-order polynomial is used for planning, given the initial position and velocity r1. End point position and velocity r4 The oscillation trajectory of the fifth-order polynomial can be solved by finding the midpoints r2 and r3, i.e., r = k5t. 5 +k4t 4 +k3t 3 +k2t2 +k1t+k0,t∈(0,T), where r2 coincides with the edge corner of the virtual obstacle, and r3 is the highest point of the swing.

[0115] Assuming the obstacle is directly in front of the robot, r4 is calculated as follows:

[0116]

[0117]

[0118] Where, r hip_x d1 represents the position of the hip joint in the x-direction relative to the origin, and d2 represents the safety distance.

[0119] In steps S150 and S160 above, as Figure 6 and Figure 7 As shown, when establishing the tracking control model for the swing leg and supporting leg, the online tracking control model for the swing leg trajectory is first established:

[0120] θ d =IK(r)

[0121] Where IK represents the inverse kinematics of the leg, θ d The desired joint angle vector;

[0122] Then, the target trajectory of the machine is input, and the supporting leg foot force is output for online rolling optimization.

[0123] The dynamics of the online robot are described as follows:

[0124]

[0125] in,

[0126]

[0127] The above formula can be written as:

[0128] x′ k+1 =A′x′ k +Bf k +G k

[0129] The minimum cost function min f′(x) is:

[0130]

[0131] Among them, Q′ x and R′ u Weight matrix; For the trajectory of the target machine; u′ k The desired input is x′; T is the oscillation period; N is the number of iterations;k Let u′ represent the current trajectory of the machine, where u′ = [F1, F2, F3, F4]. T .

[0132] The constraints of the aforementioned minimum cost function min f′(x) include at least the system dynamics constraints, force constraints, and friction cone constraints of the quadruped robot.

[0133] Specifically, the constraints are:

[0134] stx′ k+1 =A′x′ k +Bf k +G k System Dynamics

[0135] Force Constraint

[0136] Friction cone constraint

[0137] Furthermore, using the optimized foot force as input, a joint compliance servo control model is established, which is as follows:

[0138]

[0139] In the formula: K pθ K is the controller stiffness coefficient matrix; vθ θ is the controller damping coefficient matrix; θ is the actual joint angle vector. This represents the desired joint angular velocity vector; τ is the actual joint angular velocity vector. ff =-J T u′ k is the joint force feedforward term; J is the joint force Jacobian matrix; u is the joint controller input parameter.

[0140] In the above specific embodiments, the adaptive jumping control method for quadruped robots provided by the present invention acquires the elevation information of the target obstacle and calculates the target pose of the robot trajectory based on the elevation information; solves a pre-constructed target optimization function according to set constraints to obtain the target motion trajectory of the robot jumping over the target obstacle; generates a safe landing position for the swing leg based on the current position and the target position of the swing leg; calculates the swing trajectory of the swing leg based on the safe landing position and the target motion trajectory; constructs a trajectory tracking model with the target motion trajectory as input, solves the trajectory tracking model with a minimum cost function to obtain the optimized foot force of the supporting leg; constructs a joint compliant servo control model with the optimized foot force of the supporting leg as input, and calculates the joint motion parameters under the target pose of the robot trajectory through the joint compliant servo control model, so that the control device of the legged robot outputs control commands according to the joint motion parameters, and the control commands are used to control the leg joint movements of the legged robot.

[0141] This adaptive jumping control method for quadruped robots fully considers obstacle information and optimizes the generation of the robot's motion trajectory, thereby effectively enabling the robot to adapt to jumping. Simultaneously, by combining safe and unsafe zones within the robot's flight space for online planning of the swing leg's landing and swing trajectory, obstacles can be effectively avoided. This allows for autonomous and stable obstacle crossing even with larger obstacles, offering high safety and high passage efficiency. By fusing obstacle information, this method controls the robot to jump autonomously, achieving precise dynamic obstacle crossing and significantly improving the robot's passage efficiency and stability with large obstacles. It solves the problems of poor stability and high tipping risk when robots cross large obstacles, thereby improving the landing stability of legged robots when crossing obstacles, reducing tipping risk, and ultimately improving the performance of legged robots in complex terrain.

[0142] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An adaptive jumping control method for a quadruped robot, characterized in that, include: Obtain the elevation information of the target obstacle, and calculate the target pose of the computer body trajectory based on the elevation information; Based on the set constraints, the pre-constructed target optimization function is solved to obtain the target motion trajectory of the body jumping over the target obstacle; Based on the current position and target position of the swing leg, generate a safe landing position for the swing leg; Calculate the swing trajectory of the swing leg based on the safe landing position and the target movement trajectory; A trajectory tracking model is constructed using the target motion trajectory as input, and the trajectory tracking model is solved using the minimum cost function to obtain the optimized supporting leg foot end force; Using the optimized supporting leg foot force as input, a joint compliant servo control model is constructed, and the joint motion parameters under the target pose of the computer body trajectory are calculated through the joint compliant servo control model, so that the control device of the leg robot outputs control commands according to the joint motion parameters. The control commands are used to control the leg joint movements of the leg robot. Based on the current position and target position of the swing leg, a safe landing position for the swing leg is generated, specifically including: Get the current position r of the swing leg c and the target position r of the swing leg f ; Using the second expression, a linear interpolation is performed between the current position rc and the target position rf to obtain the safe landing position; The second expression is: ； in, For the swing time, For the oscillation period, ; The minimum cost function for: ； in, and Weight matrix; For the trajectory of the target machine; For the expected input; The number of iterations; This represents the current trajectory of the machine.

2. The adaptive jumping control method for a quadruped robot according to claim 1, characterized in that, Acquiring the elevation information of the target obstacle and calculating the target pose of the computer body trajectory based on the elevation information, specifically includes: Obtain the height information of the target obstacle and the distance information of the target obstacle relative to the origin; The position of the target aircraft is calculated based on the height information, the distance information, and the parameter information of the target aircraft. The trajectory and pose of the target body are calculated based on the position, attitude, average velocity, and angular velocity of the target body.

3. The adaptive jumping control method for a quadruped robot according to claim 2, characterized in that, The parameter information of the target body includes the safe distance of the target body's center of mass relative to the edge of the target obstacle, the length of the target body, and the width of the target body; Using the first expression, the position of the target aircraft is calculated based on the height information, the distance information, and the parameter information of the target aircraft. The first expression is: ； Where, p e d represents the location of the target machine. x The distance information is represented by its x-axis component, d y Let d be the y-component of the distance information. m The distance between the center of mass of the target body and the edge of the target obstacle is denoted as l, where l is the length of the target body and w is the width of the target body.

4. The adaptive jumping control method for a quadruped robot according to claim 1, characterized in that, The pre-constructed objective optimization function is: ； in, and This is the weight matrix. For the desired dynamic state of the quadruped robot system, This represents the state of the quadruped robot system in its current pose. For the desired control input, For the current control input, Number of iterations.

5. The adaptive jumping control method for a quadruped robot according to claim 4, characterized in that, The constraints set include environmental constraints and dynamic constraints; The constraints include at least the system dynamics constraints, force constraints, friction cone constraints, foot position constraints, leg length constraints, acceleration constraints, torque constraints, and joint angle constraints of the quadruped robot.

6. The adaptive jumping control method for a quadruped robot according to claim 1, characterized in that, Based on the safe landing position and the target movement trajectory, the swing trajectory of the swing leg is calculated, specifically including: Before the aircraft lands, the starting point of the swing leg is determined based on the safe landing position and the target motion trajectory. Starting point velocity End point position End point velocity The position coinciding with the edge corner of the target obstacle and the highest point of swing Construct a fifth-degree polynomial; The oscillating trajectory is obtained by solving a fifth-degree polynomial.

7. The adaptive jumping control method for a quadruped robot according to claim 1, characterized in that, The minimum cost function The constraints include at least the system dynamics constraints, force constraints, and friction cone constraints of the quadruped robot.

8. The adaptive jumping control method for a quadruped robot according to claim 1, characterized in that, The joint compliant servo control model is as follows: ； In the formula: This is the controller stiffness coefficient matrix; This is the controller damping coefficient matrix; This is the actual joint angle vector; This represents the desired joint angular velocity vector; This is the actual joint angular velocity vector; This is a joint force feedforward term; The joint force Jacobian matrix; Input parameters for the joint controller.

Images