Four-wheel-foot robot foot type motion gait planning method and four-wheel-foot robot

Abstract

The present application belongs to the technical field of mobile robots, and discloses a four-wheel-foot robot foot type motion gait planning method and a four-wheel-foot robot, wherein the method comprises the following steps: S1, in the swing period of a swing leg, the real-time center of gravity position of the robot body is calculated and obtained; S2, under the condition that the motion trajectory of the robot body is unchanged, the support position of the support leg is adjusted through the rolling of the wheels according to the real-time center of gravity position, so that the projection of the real-time center of gravity position on the ground is located in the support triangle formed by the support position of the support leg. The present application considers that when one leg of the four-wheel-foot robot swings, the wheels at the bottom of the remaining three support legs can provide support force to the robot body while moving, and based on this, it is proposed that when the four-wheel-foot robot travels in the foot type motion mode, the wheel foot in the support phase adjusts the support position through the rolling of the wheels, so as to expand the motion stability domain, which is beneficial to improve the terrain adaptability and obstacle crossing performance of the robot.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of mobile robots, and more particularly to a four-wheel-foot robot foot type motion gait planning method and a four-wheel-foot robot. BACKGROUND

[0002] The four-wheel-foot composite robot uses the leg type structure as the suspension of the wheeled mobile mechanism, and uses the wheels as the foot end of the leg type mobile mechanism, thereby integrating the wheeled mobile mechanism and the leg type mobile mechanism. Since it has the respective advantages of the wheeled robot and the leg type robot, under the condition of having road support, the wheeled mobile mechanism is used to travel, the moving speed is high, and the energy efficiency is high; and under the condition of no road support, the leg type mobile mechanism is used to walk, the appropriate foot landing points are selected in the terrain, and the robot is walked relying on these discrete foot landing points, thereby overcoming the terrain obstacles and enhancing the terrain passability of the robot. Therefore, the four-wheel-foot composite robot is considered as a "two fulls" mobile solution, and has become the focus of development in the field of mobile robots.

[0003] However, there are many deficiencies in the motion planning and control of the existing four-wheel-foot composite robot, which are mainly shown as follows: in the wheeled motion, the leg type structure is often locked and only used as a vehicle suspension structure; and in the leg type motion, the rotating shaft of the wheel is locked and the wheel is simply used as a foot end. This approach does not fully play the composite characteristics of the leg type structure and the wheeled structure, especially in the process of moving the robot in the leg type walking mode, the wheel is locked, which means that the four-wheel-foot composite robot simply becomes a four-legged robot, thereby causing the disadvantages of insufficient stable domain of the four-legged robot in the walking process, which is inevitable in the four-wheel-foot composite robot, and greatly limits the motion speed and the diversity of the motion mode of the four-wheel-foot composite robot.

[0004] When the four-legged robot walks in the low-speed static gait, a single leg walks forward in turn, and the remaining three legs are in the landing state to support and propel the robot body. After the current walking leg completes the swing and lands, the next leg in turn starts to enter the walking state, and so on, so that the robot realizes continuous walking. In this process, in order to ensure the stability of the robot, the projection of the center of gravity of the robot on the ground needs to be in the support triangle composed of the landing points of the three supporting legs, and the minimum value of the distance of the center of gravity projected to the three edges of the triangle is taken as the measure of the stability margin. However, the support triangle does not change with the movement of the leg, and if the planning is not proper, the projection of the center of gravity of the robot on the ground may be inside the triangle at the previous time, stably supporting the robot, but at the next time, the center of gravity may be very close to the edge of the triangle, or even outside the triangle, causing the robot to lose stability. Especially when the length-width ratio of the robot is large, it is more likely to deviate from the stable region when walking alternately.

[0005] The prior art quadruped robot has the disadvantage of insufficient stable domain during walking, which greatly limits the terrain adaptability and obstacle crossing performance of the robot. SUMMARY

[0006] In view of the above defects or improvement needs of the prior art, the present application provides a quadruped robot foot type motion gait planning method and a quadruped robot, which solves the problem of insufficient stable domain during walking of the prior art quadruped robot, greatly limits the terrain adaptability and obstacle crossing performance of the robot, and proposes to adjust the wheel-foot support position in the support phase through the rolling of the wheels so as to expand the motion stable domain, which is beneficial to improve the terrain adaptability and obstacle crossing performance of the robot.

[0007] To achieve the above object, according to one aspect of the present application, a quadruped robot foot type motion gait planning method is provided, comprising:

[0008] S1. In the swing period of the swing leg, the real-time center of gravity position of the robot body is calculated and obtained;

[0009] S2. In the case that the motion trajectory of the robot body is unchanged, the support position of the support leg is adjusted according to the real-time center of gravity position through the rolling of the wheels, so that the projection of the real-time center of gravity position on the ground is located inside the support triangle formed by the support position of the support leg.

[0010] According to another aspect of the present application, a quadruped robot is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the quadruped robot foot type motion gait planning method of any one of the above aspects when executing the computer program.

[0011] Overall, compared with the prior art, the quadruped robot foot type motion gait planning method and the quadruped robot provided by the present application have the following advantages:

[0012] 1. Considering that the remaining three support legs of the prior art foot type robot cannot adjust the position of the foot point due to the need for support when one leg swings, and the wheels at the bottom of the remaining three support legs of the quadruped robot can provide support force to the body while moving when one leg swings, based on this, it is proposed that when the quadruped robot travels in the foot type motion mode, the wheel-foot in the support phase adjusts the support position through the rolling of the wheels so as to expand the motion stable domain, which is beneficial to improve the terrain adaptability and obstacle crossing performance of the robot;

[0013] 2. The optimal support triangle at any time of the swing leg swing period is obtained by calculation, the support positions of the three support legs are adjusted according to the optimal support triangle, so that the three support legs form the optimal support point, at which the stability margin of the robot is maximum, so that the optimal stability performance of the robot is obtained, which is beneficial to improve the marching speed and is beneficial to make the four-wheel-foot robot have more stable and fast passability in the rugged terrain environment. BRIEF DESCRIPTION OF DRAWINGS

[0014] Figure 1 is the mechanism diagram of the four-wheel-foot robot without a side swing joint provided by the present application;

[0015] Figure 2 is the mechanism diagram of the four-wheel-foot robot with a side swing joint provided by the present application;

[0016] Figure 3 is the single leg angle definition diagram of the four-wheel-foot robot provided by the present application;

[0017] Figure 4 is the foot end trajectory curve in the specific example provided by the present application;

[0018] Figure 5 is the displacement curve of the foot end in the X direction in the specific example provided by the present application;

[0019] Figure 6 is the displacement curve of the foot end in the Z direction in the specific example provided by the present application;

[0020] Figure 7 is the optimal support triangle diagram corresponding to the left front leg of the robot when leaving the ground at the point 1 and the point 2 respectively provided by the present application;

[0021] Figure 8 is the optimal support point change trend diagram of the robot in the left front leg swing period provided by the present application;

[0022] Figure 9 is one of the optimal support triangle change trend diagrams of the robot in the left front leg swing period provided by the present application;

[0023] Figure 10 is the second optimal support triangle change trend diagram of the robot in the left front leg swing period provided by the present application;

[0024] Figure 11 is the stability margin change trend diagram of the robot in the left front leg swing period provided by the present application;

[0025] Figure 12 is the optimal support triangle diagram obtained by the robot in the process of case (b) provided by the present application;

[0026] Figure 13 is a trend chart of the optimal support triangle of the robot in the process of case (b) provided by the present application;

[0027] Figure 14 is a trend chart of the stability margin of the robot in the process of case (b) provided by the present application;

[0028] Figure 15 is a moving gait chart of the four-wheel-foot robot provided by the present application;

[0029] Figure 16 is a trend chart of the optimal support triangle of the robot provided by the present application;

[0030] Figure 17 is a schematic diagram of the change of the optimal support triangle of the robot moving between the points A and B provided by the present application;

[0031] Figure 18 is a step-over gait chart of the four-wheel-foot robot provided by the present application. DETAILED DESCRIPTION

[0032] In order to make the objectives, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and should not be used to limit the present application. In addition, the technical features involved in each embodiment of the present application described below can be combined with each other as long as they do not conflict with each other.

[0033] Please refer to Figure 1 and Figure 2 The present application provides a four-wheel-foot robot foot-type motion gait planning method, which comprises the following steps:

[0034] S1, in the swing period of the swing leg, the real-time center of gravity position of the robot body is calculated and obtained;

[0035] S2, under the condition that the motion trajectory of the body is unchanged, the support position of the support leg is adjusted by the rolling of the wheel according to the real-time center of gravity position, so that the projection of the real-time center of gravity position on the ground is located inside the support triangle formed by the support position of the support leg.

[0036] The four-wheel-foot robot foot type motion gait planning method provided by the application considers that when an existing foot type robot swings one leg, the remaining three supporting legs cannot adjust the position of the foot point due to the need for support, while when the four-wheel-foot robot swings one leg, the wheels at the bottom of the remaining three supporting legs can provide support to the body while moving, based on which, when the four-wheel-foot robot travels in a foot type motion mode, the wheel foot in the supporting phase adjusts the supporting position through the rolling of the wheels, so as to expand the motion stability domain, which is beneficial to improve the terrain adaptability and obstacle crossing performance of the robot, and is also beneficial to improve the travel speed of the robot.

[0037] Specifically, when the four-wheel-foot robot travels in a foot type motion mode, a quasi-static walking gait is adopted; the supporting position of the supporting leg is adjusted through the rolling of the wheels without affecting the original preset motion trajectory of the body, so that the real-time center of gravity position of the body in the swing leg swing phase state, i.e. in the swing period, is always located inside the supporting triangle formed by the three supporting positions on the plane of the three supporting positions, i.e. the projection of the ground, so as to improve the motion stability.

[0038] Further, S2 further comprises:

[0039] According to the real-time center of gravity position, the supporting position of the supporting leg is adjusted through the rolling of the wheels, so that the supporting position of the supporting leg forms an optimal supporting triangle, wherein the optimal supporting triangle satisfies that the projection of the real-time center of gravity position on the ground is located inside the optimal supporting triangle, the distance between the optimal supporting triangles of two adjacent moments is less than a preset distance, and the stability margin under the optimal supporting triangle is maximum.

[0040] That is, the optimal supporting triangle at any moment in the swing leg swing period can be obtained through calculation, and the supporting positions of the three supporting legs are adjusted according to the optimal supporting triangle, so that the three supporting legs form an optimal supporting point, and the stability margin of the robot is maximum at the optimal supporting point, so that the optimal stability performance of the robot can be obtained, which is beneficial to improve the travel speed and is beneficial to make the four-wheel-foot robot have more stable and fast passability in a rugged terrain environment.

[0041] Further, S1 specifically comprises:

[0042] planning to determine the foot end motion trajectory of the swing leg;

[0043] According to the foot end motion trajectory, the real-time center of gravity position is calculated and obtained in combination with the forward and inverse kinematics relationship of the swing leg.

[0044] The following is a detailed description of real-time center of gravity position acquisition: Four-wheel-foot robots are usually designed based on bionics principles, and the leg mechanism is at least composed of two serial links, i.e. thigh and shank. At the proximal end of the thigh, the thigh is connected to the robot body through the hip joint, i.e. the thigh joint hinge point, and the distal end of the thigh is connected to the proximal end of the shank through the knee joint, i.e. the shank joint hinge point. The wheel hub motor independently driven wheel is installed at the distal end of the shank, and the rotating shaft of the wheel is the ankle joint of the leg mechanism. Thus, the leg mechanism has at least three active rotary joints, i.e. hip flexion joint (HFE), knee joint (KFE), and ankle joint (FKE), as shown in Figure 1 .

[0045] Sometimes, in order to improve the flexibility of the four-wheel-foot composite robot in turning and resisting external lateral impact during foot movement, a lateral swing joint (HAA) is added to the hip to realize the lateral swing / medial movement of the leg. The leg mechanism can also be composed of multiple segments, as shown in Figure 2 .

[0046] The relationship between the single leg lateral swing joint angle θ0, the single leg thigh joint angle θ1, the single leg shank joint angle θ2, and the single leg wheel joint angle θ3 is shown in Figure 3 .

[0047] The robot-related variables are shown in Table 1.

[0048] Table 1 Robot Parameter Table

[0049]

[0050]

[0051] This paper takes the left front leg as an example to explain the algorithm solving process. First, plan the foot end trajectory in the world coordinate system of the swinging leg, and select the swing phase foot end trajectory as follows:

[0052] x(t) = a1t 3 +a2t 4 +a3t 5 ;

[0053] y(t) = 0;

[0054] z(t) = b1t 3 +b2t 4 +b3t 5 +b4t 6 ;

[0055] 0 < t < T m ;

[0056]

[0057]

[0058] The single-step duration of the robot is obtained by dividing the step length by the speed. Assuming that the duty cycle of the robot is 0.75, the step length is 0.4 m, the single-step duration is 0.7 s, the duty cycle is 0.75, the leg-lifting height is 0.16 m, and the standing height is 0.5 m, the speed of the robot is 0.571 m / s, and the swing phase period is 0.175 s. The foot trajectory in the world coordinate system is shown in FIG. 8; the X-direction displacement curve and the Z-direction displacement curve are shown in FIGS. 9 and 10, respectively. Figure 4 Figure 5 Figure 6

[0059] The body coordinate system is established as shown in FIG. 11 or 12. The world coordinate system is established on the ground plane, and the body coordinate system takes the geometric center of the body as the origin, the forward direction as the positive direction of the X axis, and the vertical upward direction as the positive direction of the Z axis, and the Y axis direction is determined according to the right-hand rule. Figure 1

[0060] This paper takes the left front leg as an example to illustrate the solving process of the algorithm and analyze the forward and inverse kinematics of the single leg of the robot.

[0061] The forward kinematics relationship is:

[0062] z(t) = -sinθ LF1 (t)L LF1 -sin(θ LF2 (t)-θ LF1 (t))L LF2 ;

[0063] x(t) = -cosθ LF1 (t)L LF1 +cos(θ LF2 (t)-θ LF1 (t))L LF2 ;

[0064] The inverse kinematics relationship is:

[0065]

[0066]

[0067] θ LF3 (t) = θ LF2 (t)-θ LF1 (t);

[0068] wherein x(t) and z(t) are the coordinates of the wheel center relative to the thigh joint coordinate system as shown in FIG. 13. In the following calculation of the real-time center of gravity position, they need to be converted into the coordinates in the body coordinate system for calculation. Figure 3

[0069] ​​​​​Similarly, the other three legs are solved. The relevant calculation formula of the robot when the other legs are stepped out is consistent with the left front leg, and the calculation formula of the four legs is the same, only need to replace the subscript LF representing the left front leg in the formula with the subscript LH representing the left rear leg, the subscript RF representing the right front leg and the subscript RH representing the right rear leg.

[0070] Further analyze the center of gravity of the robot at this time. The coordinates of the hinge point of the left front thigh joint in the body coordinate system are (x LF ,y LF ,z LF ), the hinge points of the front and rear legs are symmetrical with the y axis, and the hinge points of the left and right legs are symmetrical with the x axis. Assuming l 1x (t) is the projection of the distance between the mass center of the thigh and the hinge point in the x direction as a function of time, l 2x (t) is the projection of the distance between the mass center of the calf and the hinge point in the x direction as a function of time, l 1y (t) is the projection of the distance between the mass center of the thigh and the hinge point in the y direction as a function of time, l 2y (t) is the projection of the distance between the mass center of the calf and the hinge point in the y direction as a function of time, l 1z (t) is the projection of the distance between the mass center of the thigh and the hinge point in the z direction as a function of time, l 2z (t) is the projection of the distance between the mass center of the calf and the hinge point in the z direction as a function of time, LF indicating the left front leg. θ1(t), θ2(t) are respectively Figure 3 the functions of time. The relationship between the real-time center of gravity position and time of the robot is as follows, wherein the relationship between the coordinates of the real-time center of gravity position in the x direction and time is as follows:

[0071] COM x (t)=

[0072] [m LF1 l LF1x (t)+m LF2 l LF2x (t)+m LF3 l LF2x (t)+m RF1 l RF1x (t)+…+m LH3 l LH3x (t)] / m sum ;

[0073] In the formula:

[0074]

[0075]

[0076] l LF3x (t) = x LF - (L LF1 cos θ LF1 (t) + L LF2 cos (θ LF2 (t) - θ LF1 (t) ) ) ;

[0077] m sum = m LF1 + m LF2 + m LF3 + … + m RF1 + m LH3 + m body ;

[0078] m body is the mass of the robot body; m sum is the total mass of the robot;

[0079] The relationship between the coordinate of the real-time center of gravity position in the y direction and time is as follows:

[0080] COM y (t) =

[0081] [ m LF1 l LF1y (t) + m LF2 l LF2y (t) + m LF3 l LF2y (t) + m RF1 l RF1y (t) + … + m LH3 l LH3y (t) ] / m sum ;

[0082] In the formula:

[0083]

[0084]

[0085] l LF3y (t) = y LF + L LF1 sin θ LF0 (t) + L LF2 sin θ LF0 (t) ;

[0086] θ LF0 (t) is the relationship between the left front leg side swing joint angle and time;

[0087] The relationship between the coordinate of the real-time center of gravity position in the z direction and time is as follows:

[0088] COM z (t) =

[0089] [m LF1 l LF1z (t) + m LF2 l LF2z (t) + m LF3 l LF2z (t) + m RF1 l RF1z (t) + … + m LH3 l LH3z (t)] / m sum ;

[0090] In the formula:

[0091]

[0092]

[0093] l LF3z (t) = z LF - (L LF1 sinθ LF1 (t) + L LF2 sin(θ LF2 (t) - θ LF1 (t))).

[0094] Further, the projection of the real-time center of gravity position on the ground in S2 is specifically:

[0095] According to the three foot points of the robot, a plane passing through the three foot points is determined, that is, according to the support positions of the three support legs of the robot, a plane passing through the three support positions is determined. When the left front leg of the robot is in the swing phase state, the spatial coordinates of the left rear leg foot end, the right front leg foot end and the right rear leg foot end are A(x LH , y LH , z LH ), B(x RF , y RF , z RF ), C(x RH , y RH , z RH ) respectively. Then:

[0096]

[0097]

[0098] Assume unit variable: And:

[0099]

[0100] The analytical expression of the plane formed by the support positions of the three support legs of the robot in the body coordinate system is:

[0101]

[0102] The projection point of the real-time barycenter position on the flat ground is M'; the pitch angle a, the roll angle b and the yaw angle g of the body are obtained based on the inertial measurement unit; when the robot is on the slope, the coordinates of the projection point of the real-time barycenter position of the robot are:

[0103]

[0104] Using the same method, when the other foot wheels enter the swing phase, the coordinates of the projection point of the robot mass center are calculated according to the support triangle.

[0105] It should be pointed out that this paper only takes the left front leg as an example, and does not limit the invention, and any modification within the spirit and principles of the invention should be included in the protection scope of the invention.

[0106] During the forward movement of the quadruped robot, the mass center, i.e. the barycenter, approaches the edge of the support triangle, and sometimes even exceeds the support triangle, entering an unstable state, which is more obvious when passing over a step or a ditch. At this time, the support triangle can be optimized by adjusting the positions of the remaining three legs to improve the stability of the robot. Therefore, this paper designs a gait planning optimization algorithm to analyze the optimal support triangle in real time by taking advantage of the feature that the wheeled mechanism can provide support force to the body during movement.

[0107] Further, the acquisition of the optimal support triangle in S2 is specifically:

[0108] According to the physical parameters of the robot, the reachable foot point range of each foot of the robot is obtained;

[0109] At any time during the swing period of the swing leg, the boundaries of the reachable foot point range of each foot are traversed, the real-time barycenter position at the current time is calculated to obtain the support triangle with the maximum stability margin, and it is judged whether the projection of the real-time barycenter position on the ground is located inside the support triangle with the maximum stability margin and whether the distance from the optimal support triangle at the last time is within a preset distance;

[0110] The support triangle with the maximum stability margin is output as the optimal support triangle when the projection of the real-time barycenter position on the ground is inside, the distance from the optimal support triangle at the last time is less than the preset distance, and the projection of the real-time barycenter position on the ground is inside.

[0111] Specifically, the robot gait trajectory optimization algorithm can be expressed as follows:

[0112] [Four-wheel-foot robot gait trajectory optimization algorithm]: input: the motion trajectory of the center of mass relative to the world coordinate system, i.e. the preset motion trajectory of the body, the size parameters of the robot, the mass parameters of the robot;

[0113] (1) When a new gait cycle starts, the motion trajectory of each foot of the robot relative to the robot body coordinate system is calculated according to the motion trajectory of the center of mass of the robot relative to the world coordinate system and the relevant geometric and physical parameters of the robot, including the size, mass, etc. of the relevant structures, and the projection trajectory of the center of mass of the robot on the support surface is further calculated.

[0114] (2) When the robot is converted from a four-foot support state to a three-foot support state, it is determined whether the robot is in a stable state after the leg that is replaced into the swing phase is lifted, and whether the forward propulsion process of the body is in a quasi-static stable state under the three-foot support state. There are two cases, which are processed as follows:

[0115] (a) If the robot will be in an unstable state at the start of the swing cycle after the wheel leg that is about to swing is lifted, the support point position of the wheel foot in the support phase is changed by rotating the wheel to ensure that the robot is in a stable state during the upcoming motion process, while keeping the body position unchanged.

[0116] (b) If the support points of the remaining three wheel legs in the support phase are not sufficient to ensure the force balance condition required for the forward propulsion of the robot body during the swing of the leg in the swing phase, i.e. from the next moment of the start of the swing cycle to the end of the swing cycle, the wheel feet of the three legs in the support phase are adjusted synchronously according to the optimal support point change trajectory to achieve the condition that the robot is always in force balance during motion, until the leg swing ends, while keeping the body motion trajectory unchanged.

[0117] The optimal support point is calculated by the following steps: within the current reachable footfall range of each support leg, a polygon is determined, and the center of mass of the robot body is within the polygon, the distance from the selected polygon at the last moment is limited, and the stability margin is maximum. The distance from the selected polygon at the last moment is limited, i.e. if the distance is too large, the robot foot cannot move to the specified position. When the robot is in the state before starting, the points of the polygon are the optimal support points. When the robot is in the state during movement, the optimal motion trajectory of each foot in the support phase is obtained according to the polygon at different times.

[0118] Further, the stability margin is the minimum distance between the projection of the real-time center of gravity on the ground and the three edges of the support triangle, or the minimum distance between the projection of the real-time center of gravity on the ground and the three edges of the support triangle in the direction of robot movement. That is, the embodiment proposes two calculation methods of the stability margin, and both of the stability margins can be used as the basis for calculating the optimal support triangle in the swing period, and the specific method is not limited.

[0119] Further, the embodiment divides the swing period of the swing leg into the starting time and the process from the next time of the starting time to the end of the swing period, that is, two cases of (a) and (b) mentioned above. In a specific embodiment, both cases use the minimum distance between the projection of the real-time center of gravity on the ground and the three edges of the support triangle as the stability margin to obtain the optimal support triangle. Specifically, for case (a) mentioned above, the specific calculation steps of the optimal support point are as follows:

[0120] First step: input the size parameters of the robot and the standing height, and output the reachable foot point range of each foot of the robot;

[0121] Second step: for case (a) mentioned above: traverse the boundary of the reachable foot point range of each foot, calculate the distances l1, l2, l3 of the edges of the planar triangle formed by the three support points to the center of mass, and l x = min (l1, l2, l3); if l x is greater than the maximum value currently calculated, further judge whether the center of mass is in the triangle formed by the three edges. If all conditions are met, l = l x After traversal, update the matrix coordinates of the three-dimensional coordinates of the three support points.

[0122] Third step: output the coordinate matrix to the main control program, and the main control program calculates the motor angle by the forward kinematics method according to the target three-dimensional coordinates of the optimal support point and sends it to the driver.

[0123] For case (b) mentioned above, the specific calculation steps of the optimal support point are as follows:

[0124] First step: input the size parameters of the robot and the standing height, and output the reachable foot point range of each foot of the robot, and take t = 0;

[0125] Second step: traverse the boundary of the reachable foot point range of each foot, calculate the distances l1, l2, l3 of the edges of the planar triangle formed by the three support points to the center of mass, and l x = min (l1, l2, l3); if l xIf the distance is greater than the maximum value calculated at present, it is further judged whether the center of mass is in the triangle formed by the three edges. If it is in the triangle, it is judged whether the distance from the last best point is greater than a preset distance, for example, 100 mm. If all the conditions are met, l = l x , the three support point coordinates are recorded and the matrix recording the three-dimensional coordinates of the three support points is updated, t = t + 0.01. Here, 0.01 is the time interval for obtaining the best support triangle, and can also be other values, which are not limited in particular.

[0126] Step 3: If t < T m Return to Step 2.

[0127] Step 4: Output the matrix output coordinates to the main control program, and the main control program calculates the motor rotation angle through the forward kinematics method according to the three-dimensional coordinates of the support points at each moment and sends it to the driver.

[0128] With the four legs of the robot entering the swing phase or the support phase in turn, the motion of the other legs in the support phase is analyzed and calculated according to the above method, to ensure the stable movement of the robot.

[0129] The judgment of whether the center of mass is in the triangle is because if only the stability margin is calculated, there is a case that the support triangle is outside the center of mass, and the minimum distance of the three edges from the center of mass is the maximum. The judgment of the last best point is because the distance is limited by the speed of the robot foot end, that is, if the distance is too large, the robot foot end cannot move to the specified position. The support leg can be moved to the best support point by controlling the side swing of the support leg, the angle of the thigh joint and the angle of the shank joint.

[0130] Further, the stability margin is specifically the minimum value of the projection of the real-time center of gravity position on the ground and the distance between the three edges of the support triangle. For example, when the left front leg of the robot is in the swing phase, the coordinates of the left rear leg foot end, the right front leg foot end and the right rear leg foot end in the space coordinate system with the center of mass as the origin, that is, in the body coordinate system, are A(x LH , y LH , z LH ), B(x RF , y RF , z RF ), and C(x RH , y RH , z RH ), respectively. The distances of the projection of the real-time center of gravity position on the ground from the edges AB, BC and AC of the triangle are respectively:

[0131]

[0132]

[0133]

[0134] The stability margin δ of the robot at this time is the minimum of the three:

[0135] δ = min(d M″AB , d M″BC , d M″AC ).

[0136] Taking the parameters in Table 2 as an example, for case (a), the analysis result is as shown in Figure 7 .

[0137] Based on the joint angle of the IMU data and motor feedback, the body state is estimated. The robot size, weight and other parameters are shown in Table 2.

[0138] Table 2 Robot size parameter table

[0139] Thigh leg length 350 mm Calf leg length 350 mm Body length 440 mm Body width 300 mm Body thickness 100 mm Wheel diameter 150 mm Single leg mass 6.43 kg Machine body mass 8.75 kg

[0140] Assuming Figure 7 the left middle point is the origin point 1, and the left upper point is the origin point 2. After algorithm calculation, the best support triangle corresponding to the origin point 1 is obtained as shown by the dashed line, and the best support triangle corresponding to the origin point 2 is obtained as shown by the solid line.

[0141] From Figure 7 it can be seen that if the robot leg is not adjusted, during the process of single leg stepping, the center of mass has exceeded the support triangle, the robot has fallen forward, and enters an unstable state, so that the left front foot touches the ground in advance. In addition, it can be obtained that the best triangle is all isosceles triangle. This is because at this time, when the support point improves the distance between the support triangle edge and the center of mass, it will inevitably lead to the reduction of the distance of the other edge.

[0142] Based on the specific parameters in Table 2, from the start time of the left front leg swing to the end of the swing, the projection point coordinates of the center of mass change from (0, 18) to (0, 186). When the center of mass moves from (0, 18) to (0, 186) during the left front leg swing period, the coordinate trajectories of the corresponding three best support points are as shown in Figure 8 .

[0143] During the swing period, the process of moving the center of mass is divided into 75 segments, that is, 75 time points are uniformly selected, and the three support points are connected to obtain the projection of the support triangle on the ground with time, as shown in Figure 9 .

[0144] Further processing, the support triangle change diagram is obtained as shown in Figure 10As shown in the figure, where the z-axis represents the distance between the robot centroid and the body centroid. That is, according to this kind of stability margin calculation method, in the swing leg swing period, the optimal support triangle is constantly changing, so that each support leg respectively forms a real-time motion trajectory and reaches the optimal support triangle position.

[0145] Further, the stability margin trend chart when the left front leg enters the swing phase is obtained, as shown in the figure. Figure 11

[0146] As shown in the figure. Figure 11 It can be seen that the distance between the robot centroid and the support triangle boundary is the largest when the robot centroid is (0, 18), which can reach 165 mm, and the minimum distance is 130 mm when the left front foot lands. Combined with the robot standing height, according to the minimum distance calculation, it can be obtained that the robot can normally start on a slope of 14.6 degrees, and if the support triangle is not adjusted, the robot will enter an unstable state and fall forward even on flat ground.

[0147] Further, the stability margin is specifically the distance between the projection of the real-time gravity center position on the ground and the minimum value of the three edges of the support triangle in the robot advancing direction, and the stability margin is specifically:

[0148] Taking the left front leg swing as an example, when the left front leg of the robot is in the swing phase, the coordinates of the left rear leg foot end, the right front leg foot end, and the right rear leg foot end in the space coordinate system with the body centroid as the origin, i.e. the body coordinate system, are A(x LH , y LH , z LH ), B(x RF , y RF , z RF ), and C(x RH , y RH , z RH ), respectively. The distance between the projection point of the real-time gravity center position on the ground and the projection of straight line AB on the horizontal plane in the robot advancing direction is:

[0149] x M″ is the coordinate of the projection point of the real-time gravity center position on the ground in the x direction of the body coordinate system; y M″ is the coordinate of the projection point of the real-time gravity center position on the ground in the y direction of the body coordinate system.

[0150] The distance between the projection point of the real-time gravity center position on the ground and the projection of straight line AB on the sagittal plane in the robot advancing direction is:

[0151] z M″ ​the coordinate of the projection point of the real-time barycentric position on the ground in the z direction of the body coordinate system;

[0152] the distance between the projection point of the real-time barycentric position and the straight line AB in the advancing direction is:

[0153]

[0154] Similarly, the distance between the projection point of the real-time barycentric position and the horizontal projection of the straight line BC in the advancing direction of the robot is:

[0155]

[0156] the distance between the projection point of the real-time barycentric position and the sagittal projection of the straight line BC in the advancing direction of the robot is:

[0157]

[0158] the distance between the projection point of the real-time barycentric position and the straight line BC in the advancing direction is:

[0159]

[0160] Similarly, the distance between the projection point of the real-time barycentric position and the horizontal projection of the straight line AC in the advancing direction of the robot is:

[0161]

[0162] the distance between the projection point of the real-time barycentric position and the sagittal projection of the straight line AC in the advancing direction of the robot is:

[0163]

[0164] the distance between the projection point of the real-time barycentric position and the straight line AC in the advancing direction is:

[0165]

[0166] the stability margin δ of the robot at this time is the minimum value of the three:

[0167] δ = min (d M″AB , d M″BC , d M″AC ).

[0168] Further, the acquisition of the optimal support triangle in S2 is specifically:

[0169] At the beginning of the swing cycle of the swing leg, the minimum distance between the projection of the real-time center of gravity position on the ground and the three edges of the support triangle is taken as the stability margin, and the optimal support triangle is calculated and obtained, and the support position of the support leg is adjusted according to the optimal support triangle before the beginning of the swing cycle of the swing leg.

[0170] During the process from the next moment of the beginning of the swing cycle of the swing leg to the end of the swing cycle, the minimum distance between the projection of the real-time center of gravity position on the ground and the three edges of the support triangle in the direction of robot movement is taken as the stability margin, and the optimal support triangle is calculated and obtained.

[0171] That is, in the embodiment, the swing process of the swing leg is divided into two processes, namely, case (a) at the beginning of the swing cycle and case (b) during the process from the next moment of the beginning of the swing cycle to the end of the swing cycle, the minimum distance between the projection of the real-time center of gravity position on the ground and the three edges of the support triangle is taken as the stability margin in case (a). In case (b), the minimum distance between the projection of the real-time center of gravity position on the ground and the three edges of the support triangle in the direction of robot movement is taken as the stability margin; the optimal support triangle is calculated and obtained according to the stability margin, and the support position of the support leg is adjusted according to the optimal support triangle during the process of case (b).

[0172] Taking the parameters in Table 2 as an example, when the center of mass moves from (0, 18) to (0, 186), for case (b), that is, during the process from the next moment of the beginning of the swing cycle of the swing leg to the end of the swing cycle, the minimum distance between the projection of the real-time center of gravity position on the ground and the three edges of the support triangle in the direction of robot movement is taken as the stability margin to calculate and obtain the optimal support triangle, and the analysis results are shown in Figure 12 and Figure 13 The optimal support triangle remains unchanged during the process of case (b), and the coordinates of the three support points do not need to be changed during the movement process and continuously locate on the edges and corners of the reachable footfall range.

[0173] Further, the stability margin trend chart when the left front leg enters the swing phase, that is, during the process of case (b), is obtained when the support wheel foot position is planned according to this method under different center of mass positions, as shown in Figure 14

[0174] ​It can be seen from the figure that the stability margin is maximum at the centroid (0, 18) and can reach 502 mm, and is minimum at (0, 186) and is 332 mm. In combination with the height of the robot, the minimum distance is calculated, and it can be obtained that the robot can be subjected to a front and back impact of 0.66 times its own gravity without falling, and if the support triangle is not adjusted, the robot will enter an unstable state even on the flat ground and fall forward.

[0175] It can be seen that in case (b), the projection of the real-time center of gravity position on the ground and the minimum value of the distance of the three edges of the support triangle in the direction of the robot movement are used as the stability margin to calculate and obtain the optimal support triangle, and the calculation scheme is beneficial to reduce the moving frequency of the support leg, improve the stability, and simplify the control in case (b).

[0176] The robot provided by the application has a wheel-foot composite motion and a higher stability gait mode, and the front and back rolling of the foot end wheels of a single foot or multiple feet adjusts the support triangle and increases the support stability domain of the robot. The feet of the robot are alternately in a swing phase, and the remaining support feet are rolled forward and backward according to the gait planning method to generate a gait pattern as shown in Figure 15 、 Figure 18 .

[0177] The specific derivation of the subsequent robot moving out of the left rear foot, the right front foot and the right rear foot respectively obtains the gait time diagram of the wheel-foot robot as shown in Figure 15 , and the support triangle change trend diagram as shown in Figure 16 . Figure 15 In the figure, the left arrow represents backward, the right arrow represents forward, and the shaded space represents the support leg in the support phase, and the white space represents the swing leg in the swing phase. Figure 15 In the figure, the robot moves out of the left front foot, the left rear foot, the right front foot and the right rear foot in turn according to the Walk gait. During this period, the position of the remaining three support wheel legs is adjusted by using the gait planning method described in the present application, for example, in the first step, the left rear leg and the right rear leg are moved forward, and the right front leg is moved backward. It can be seen that in the process of the robot moving forward, one of the wheel-foot legs is moved backward and provides a support force, which can effectively increase the support triangle and improve the stability of the robot.

[0178] When the robot passes through the ditch, the step is large and is prone to falling. Referring to Figure 17 , points A and B are respectively set at points (-215, 520) and (35, 20) in the reachable foot landing point range. The left front leg of the robot can be selected at any point on the line segment AB to leave the ground. According to case (a), the change of the optimal support triangle of the robot corresponding to the optimal support triangle during the process of the left front leg moving from B to A is shown in Figure 17 .

[0179] After the robot crosses the ditch, the right front foot starting position can be adjusted by the rotation of the wheels. At this time, the gait diagram of the robot is as shown in Fig. 6: Figure 18 Figure 18 In the process, the robot first steps out the left front leg and stands it on the step, during which the gait planning method described in the present application is used to adjust the forward movement of the remaining three support wheel legs by different distances, then the right front leg is adjusted to a reasonable position through the pure rolling of the four tires, then the right front leg is stepped out and stood on the step, during which the gait planning method described in the present application is used to adjust the forward movement of the remaining three support wheel legs by different distances, the left rear leg is adjusted to a reasonable position through the pure rolling of the four tires again, and so on, to complete the task of climbing the step.

[0180] It is worth pointing out that the present application only takes the walk gait as an example, and the planning method result is obtained according to the parameters of a certain robot, which does not limit the present application, and any modification within the spirit and principles of the present application should be included in the protection scope of the present application.

[0181] Further, the present application also provides a four-wheel-foot robot, including a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor executes the computer program to realize the four-wheel-foot robot foot movement gait planning method according to any one of the above.

[0182] Further, the present application also provides a non-transitory computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to realize the steps of the four-wheel-foot robot foot movement gait planning method according to any one of the above embodiments.

[0183] In addition, the logical instructions in the above-mentioned memory can be realized in the form of a software functional unit and sold or used as an independent product, which can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application or parts of the present application that essentially contribute to the prior art or parts of the technical solutions can be embodied in the form of a software product, which is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in the embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

[0184] ​Those skilled in the art can easily understand that the above description is only the preferred embodiment of the present application, and is not intended to limit the present application. Any modification, equivalent replacement and improvement made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for gait planning in the legged motion of a four-wheeled legged robot, characterized in that, include: S1, during the swing cycle of the swing leg, calculate and obtain the real-time center of gravity position of the robot body; S2, with the motion trajectory of the machine unchanged, the support position of the support leg is adjusted by the rolling of the wheel according to the real-time center of gravity position, so that the projection of the real-time center of gravity position on the ground is located inside the support triangle formed by the support position of the support leg. S2 further includes: The support position of the support leg is adjusted by the rolling of the wheel according to the real-time center of gravity position, so that the support position of the support leg forms an optimal support triangle, wherein the optimal support triangle satisfies the following conditions: the projection of the real-time center of gravity position on the ground is located inside the optimal support triangle, the distance between the optimal support triangle at two adjacent moments is less than a preset distance, and the stability margin under the optimal support triangle is maximized. S1 specifically includes: The trajectory of the swing leg's foot movement is planned and determined; Based on the foot movement trajectory and the forward and inverse kinematics of the swinging leg, the real-time center of gravity position is calculated and obtained. The specific trajectory of the swing leg's foot movement described in S1 is as follows: ； ； ； ； ； ； Where S is the step size; T m The swing period of the swinging leg; Taking the left foreleg as the swinging leg as an example, the forward kinematic relationship is as follows: ； ； The inverse kinematic relationship is: ； ； ； in, The angle of the left foreleg thigh joint. The angle of the left foreleg's lower leg joint. The rotation angle of the left front wheel. Thigh length The length of the lower leg; Taking the left foreleg as the swing leg as an example, the coordinates of the thigh joint hinge point of the left foreleg in the fuselage coordinate system are (x... LF , y LF , z LF The relationship between the real-time center of gravity position in the x-direction and time is as follows: ； In the formula: ； ； ； ； The mass of the robot body; The total mass of the robot; The relationship between the real-time center of gravity position in the y-direction and time is shown in the following formula: ； In the formula: ； ； ； The relationship between the angle of the left foreleg lateral swing joint and time; The relationship between the real-time center of gravity position in the z-direction and time is shown in the following formula: ； In the formula: ； ； 。 2. The gait planning method for a four-wheeled legged robot as described in claim 1, characterized in that, The projection of the real-time center of gravity position onto the ground as described in S2 is specifically as follows: Based on the support positions of the robot's three supporting legs, the plane passing through the three support positions is determined. When the robot's left front leg is in the swing phase, the coordinates of the left hind leg foot, the right front leg foot, and the right hind leg foot in the body coordinate system are A( x LH , y LH , z LH ), B ( x RF , y RF , z RF ), C( x RH , y RH , z RH ),but: ； ； Assume a unit variable: ,and: ； The analytical expression of the plane formed by the support positions of the robot's three supporting legs in the body coordinate system is: ； Let the projection point of the real-time center of gravity on the flat ground be... Based on the pitch angle α, roll angle β, and yaw angle γ of the robot obtained by the inertial measurement unit, the coordinates of the projection point of the robot's real-time center of gravity when the robot is on a slope are: 。 3. The gait planning method for a four-wheeled legged robot as described in claim 1, characterized in that, The specific method for obtaining the optimal supporting triangle described in S2 is as follows: Based on the robot's physical parameters, obtain the range of reachable landing points for each of the robot's legs; At any moment during the swinging leg's swing cycle, the boundaries of the reachable landing points of each foot are traversed. Based on the real-time center of gravity position at the current moment, the support triangle with the largest stability margin is calculated and obtained. It is then determined whether the projection of the real-time center of gravity position on the ground is located within the support triangle with the largest stability margin and whether the distance from the optimal support triangle at the previous moment is within a preset distance. The optimal support triangle is the one whose real-time center of gravity projection on the ground is located inside the triangle, whose distance from the optimal support triangle at the previous moment is less than a preset distance, and whose stability margin is the largest.

4. The gait planning method for a four-wheeled legged robot as described in claim 1, characterized in that, The specific method for obtaining the optimal supporting triangle described in S2 is as follows: At the beginning of the swing cycle of the swing leg, the minimum value of the distance between the projection of the real-time center of gravity position on the ground and the three sides of the support triangle is used as the stability margin. The optimal support triangle is calculated and obtained. Before the swing cycle of the swing leg begins, the support position of the support leg is adjusted according to the optimal support triangle. During the period from the moment following the start of the swing leg's swing cycle to the end of the swing cycle, the minimum distance between the projection of the real-time center of gravity position on the ground and the distance between the three sides of the support triangle in the robot's travel direction is used as the stability margin to calculate and obtain the optimal support triangle.

5. The gait planning method for a four-wheeled, legged robot as described in claim 4, characterized in that, The stability margin is specifically defined as the minimum distance between the projection of the real-time center of gravity position onto the ground and the three sides of the supporting triangle. Taking the left front leg swing as an example, when the robot's left front leg is in the swing phase, the coordinates of the left hind leg foot, the right front leg foot, and the right hind leg foot in the body coordinate system are respectively A( , , ), B ( , , ), C( , , If the projection of the real-time center of gravity position onto the ground is a distance from the sides AB, BC, and AC of the supporting triangle, then the distances are respectively: ； ； ； The robot's stability margin δ at this point is the minimum of the three: 。 6. The gait planning method for a four-wheeled legged robot as described in claim 4, characterized in that, The stability margin is specifically defined as the minimum distance between the projection of the real-time center of gravity position on the ground and the distance between the three sides of the supporting triangle in the robot's travel direction. Taking the left front leg swing as an example, when the robot's left front leg is in the swing phase, the coordinates of the left hind leg foot, right front leg foot, and right hind leg foot in the body coordinate system are respectively A( , , ), B ( , , ), C( , , The distance between the projection point of the real-time center of gravity position in the robot's direction of travel and the projection of the straight line AB onto the horizontal plane is: ； The distance between the projection point of the real-time center of gravity position in the robot's travel direction and the projection of the straight line AB onto the sagittal plane is: ； The distance between the projection point of the real-time center of gravity and the straight line AB in the direction of travel is: ； Similarly, the distance between the projection point of the real-time center of gravity position in the robot's direction of travel and the projection of the line BC onto the horizontal plane is: ； The distance between the projection point of the real-time center of gravity position in the robot's travel direction and the projection of the line BC onto the sagittal plane is: ； The distance between the projection point of the real-time center of gravity and the straight line BC in the direction of travel is: ； Similarly, the distance between the projection point of the real-time center of gravity position in the robot's direction of travel and the projection of the straight line AC onto the horizontal plane is: ； The distance between the projection point of the real-time center of gravity position in the robot's travel direction and the projection of the straight line AC onto the sagittal plane is: ； The distance between the projection point of the real-time center of gravity and the straight line AC in the direction of travel is: ； The robot's stability margin δ at this point is the minimum of the three: 。 7. A four-wheeled legged robot, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the gait planning method for the four-wheeled legged robot as described in any one of claims 1-6.

Images