Multi-constraint motion planning method and system for combined octapod robots

Abstract

The present invention discloses a multi-constraint motion planning method and system for a combined octupled robot. The motion planning of the octupled robot during obstacle crossing is divided into two parts: foot-end trajectory planning for the swing phase and obstacle crossing planning. Obstacle crossing planning includes body position trajectory planning and obstacle crossing action planning. In the foot-end trajectory planning for the swing phase, by dividing the swing phase into climbing legs and non-climbing legs, a double-arc reference curve suitable for multi-constraint situations is constructed, and then a cubic uniform B-spline curve is designed as the foot-end trajectory of the climbing leg. The number of control points is adjusted to adapt to the obstacle terrain. In the obstacle crossing planning, by dividing the obstacle crossing process into a motion phase and a posture adjustment phase, multiple constraints are designed in stages, and body position trajectory planning and obstacle crossing action planning are performed to improve the ability to pass through obstacle terrain. The established multi-constraint motion planning method takes into account the influence of the obstacle terrain on the foot-end trajectory and the body posture during the obstacle crossing process, thereby improving the terrain adaptability of the combined octupled robot.

Description

Technical Field

[0001] The present invention relates to the technical field of robot motion planning, and in particular to a multi-constraint motion planning method and system for a combined octapod robot. Background Art

[0002] The statements in this section merely mention background art related to the present invention and do not necessarily constitute prior art.

[0003] Quadruped robots, due to their excellent stability, flexibility, and terrain adaptability, are widely used in various scenarios, especially in obstacle-crossing tasks. A combined octopod robot with waist pitch freedom, by connecting two quadruped robots in series, not only retains the robust motion characteristics of quadruped robots but also further enhances their obstacle-crossing capabilities. This octopod robot with waist pitch freedom has more foot support points, greatly improving its stability during obstacle-crossing. Furthermore, the introduction of waist pitch freedom greatly expands the octopod robot's motion planning methods for obstacle-crossing and enhances its flexibility.

[0004] In terms of foot trajectory planning, traditional swing-phase foot trajectory curves include polynomial curves, compound cycloids, and Bezier curves. Bezier curves are often used as foot trajectory curves during obstacle traversal because they can change the curve shape by changing the positions of control points. However, Bezier curves have some limitations when modifying the curve shape. For example, modifying a single control point may affect the entire curve, making the modification process more cumbersome. In addition, the number of control points limits the ability to flexibly represent the foot trajectory during obstacle traversal. In the more complex configuration of a modular octapod robot, swing-phase foot trajectory planning requires more advanced methods to meet the requirements.

[0005] At present, the following core problems still exist in the motion planning of the combined octapod robot during the obstacle crossing process: the increase in degrees of freedom leads to higher requirements for the complexity of the motion planning method; and higher requirements for the trajectory planning method of the swinging leg end during the obstacle crossing process. Summary of the Invention

[0006] In order to address the deficiencies of the prior art, the present invention provides a multi-constraint motion planning method and system for a combined octapod robot, taking into account both terrain adaptation and posture control.

[0007] According to one aspect of an embodiment of the present application, a multi-constraint motion planning method for a combined octapod robot is provided, the method comprising:

[0008] The combined octapod robot is divided into two front and rear quadruped robots: a first quadruped robot and a second quadruped robot;

[0009] The structured step terrain is used as the obstacle terrain, and the swing phase is divided into the climbing leg and the non-climbing leg. A double-arc reference curve is established under the multi-constraint conditions of the climbing leg foot end trajectory. The number of control points on the double-arc reference curve is set so that the control point spacing satisfies the control point spacing non-collision inequality constraint. A cubic uniform B-spline curve is obtained through the control points. The cubic uniform B-spline curve is used as the swing trajectory of the climbing leg climbing the structured step terrain, and the sine curve is used as the swing trajectory of the non-climbing leg foot end.

[0010] Multiple obstacle-crossing stages are set for the combined octapod robot; based on the multi-constraint conditions of the climbing leg foot end trajectory, multiple constraints are established for the position trajectories of the front, rear, and waist joints of the first and second quadruped robots for each obstacle-crossing stage; based on the multi-constraint conditions of the position trajectories of the front, rear, and waist joints of the first and second quadruped robots for each obstacle-crossing stage, the position trajectories of the front, rear, and waist joints of the first and second quadruped robots are obtained;

[0011] The obstacle crossing action is completed according to the swing trajectory of the climbing legs climbing the structured step terrain, the swing trajectory of the foot end of the non-climbing legs and the position trajectory of the front, rear and waist joints of the first and second quadruped robots.

[0012] Furthermore, the climbing leg foot end trajectory has multiple constraints, including: climbing distance inequality constraint, climbing state step length inequality constraint, non-climbing state step length inequality constraint, and double arc reference curve equality and inequality constraints.

[0013] Furthermore, the front part of the first quadruped robot is the front part of the octapod robot, and the rear part of the second quadruped robot is the rear part of the octapod robot; the foot-end workspaces of the first and second quadruped robots with two-degree-of-freedom single legs are established, and according to the quadruped robot structure and the foot-end workspace, the maximum step length range of each leg of the first and second quadruped robots at different body pitch angles is obtained.

[0014] According to one aspect of an embodiment of the present application, a multi-constraint motion planning system for a combined octapod robot is provided, the system comprising:

[0015] A setting module is configured to: divide the combined octapod robot into two front and rear quadruped robots: a first quadruped robot and a second quadruped robot;

[0016] A curve establishment module is configured to: take the structured step terrain as the obstacle terrain, divide the swing phase into the climbing leg and the non-climbing leg, establish a double-arc reference curve under multiple constraints of the climbing leg foot end trajectory, establish a collision-free inequality constraint on the control point spacing based on the double-arc reference curve, set the number of control points on the double-arc reference curve so that the control point spacing satisfies the collision-free inequality constraint on the control point spacing, obtain a cubic uniform B-spline curve through the control points, use the cubic uniform B-spline curve as the swing trajectory of the climbing leg climbing the structured step terrain, and use the sine curve as the swing trajectory of the non-climbing leg foot end;

[0017] The trajectory generation module is configured to: set multiple obstacle crossing stages for the combined octapod robot; establish multiple constraints for the position trajectories of the front, rear, and waist joints of the first and second quadruped robots for each obstacle crossing stage based on the multiple constraints for the trajectories of the climbing leg and foot ends; and obtain the position trajectories of the front, rear, and waist joints of the first and second quadruped robots for each obstacle crossing stage based on the multiple constraints for the position trajectories of the front, rear, and waist joints of the first and second quadruped robots.

[0018] The obstacle crossing module is configured to complete the obstacle crossing action according to the swing trajectory of the climbing leg climbing the structured step terrain, the swing trajectory of the non-climbing leg foot end and the position trajectory of the front, rear and waist joints of the first and second quadruped robots.

[0019] According to one aspect of an embodiment of the present application, a robot is provided, comprising a processor and a memory, wherein the memory stores at least one instruction, at least one program, a code set, or an instruction set, and the at least one instruction, the at least one program, the code set, or the instruction set is loaded and executed by the processor to implement the above-mentioned robot control method.

[0020] According to one aspect of an embodiment of the present application, a computer-readable storage medium is provided, wherein the storage medium stores at least one instruction, at least one program, a code set, or an instruction set, and the at least one instruction, the at least one program, the code set, or the instruction set is loaded and executed by a processor to implement the above-mentioned robot control method.

[0021] According to one aspect of an embodiment of the present application, a computer program product or computer program is provided, which includes computer instructions stored in a computer-readable storage medium. The robot's processor reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the robot executes the above-mentioned robot control method.

[0022] One of the above technical solutions has the following advantages or beneficial effects:

[0023] 1. Comprehensively consider the climbing distance inequality constraints, step length inequality constraints, and double-arc reference curve constraints when a modular octapod robot climbs structured step terrain. Under these multiple constraints, a double-arc reference curve is designed from the starting position to the end position of the climbing leg foot end. Considering the characteristics of the structured step terrain, a swing curve with a specified number of control points is further designed as the foot end trajectory of the climbing leg climbing the structured step terrain.

[0024] 2. The motion planning of the octapod robot during obstacle crossing is divided into two parts: foot-end trajectory planning for the swing phase and obstacle crossing planning. Obstacle crossing planning includes body position trajectory planning and obstacle crossing motion planning. In the foot-end trajectory planning for the swing phase, by dividing the swing phase into climbing legs and non-climbing legs, a double-arc reference curve suitable for multi-constraint situations is constructed. Then, a cubic uniform B-spline curve is designed as the foot-end trajectory of the climbing legs. The number of control points is adjusted according to the constraints to better adapt to the obstacle terrain.

[0025] 3. In obstacle planning, the robot's ability to navigate challenging terrain is enhanced by dividing the obstacle-crossing process into a motion phase and a posture adjustment phase, designing multiple constraints in each phase, and performing both body position trajectory planning and obstacle-crossing motion planning. This multi-constraint motion planning method fully considers the impact of obstacles on the leg trajectory and body posture during obstacle crossing, improving the robot's terrain adaptability. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The accompanying drawings, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0027] Figure 1 It is a flow chart of the motion planning method in the present invention.

[0028] Figure 2 This is a schematic diagram of the foot-end workspace of a quadruped robot.

[0029] Figure 3(a)-Figure 3(c) This is a schematic diagram of the foot-end workspace of a quadruped robot at different body pitch angles.

[0030] Figure 4 It is a schematic diagram of a double arc reference curve.

[0031] Figure 5 It is a schematic diagram of the shortest horizontal distance from the starting point of the climbing leg to the structured step terrain.

[0032] FIG6(a) and FIG6(b) are schematic diagrams of the double-arc reference curve passing through the extreme position of the intersection point and the terrain corner.

[0033] FIG7( a ) and FIG7 ( b ) are schematic diagrams of double-arc reference curves under double-ray constraints.

[0034] Figure 8 It is a schematic diagram of the control point positions of the B-spline curve.

[0035] Figure 9 Schematic diagram of the sinusoidal curve trajectory of the foot end of the non-climbing leg.

[0036] Figure 10(a)-Figure 10(c) This is a schematic diagram of the posture adjustment stage of the octapod robot.

[0037] Figure 11 This is a schematic diagram of a combined octapod robot with waist pitch freedom.

[0038] Figure 12 This is a top view of a modular octapod robot with waist pitch freedom. DETAILED DESCRIPTION

[0039] It should be noted that the following detailed descriptions are exemplary and intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which the present invention belongs.

[0040] Example 1: This embodiment provides a multi-constraint motion planning method for a combined octapod robot with waist pitch freedom;

[0041] like Figure 1 As shown in FIG, a multi-constraint motion planning method for a combined octapod robot includes:

[0042] S201: dividing the combined octapod robot into two front and rear quadruped robots: a first quadruped robot and a second quadruped robot;

[0043] S202: Taking the structured step terrain as the obstacle terrain, dividing the swing phase into the climbing leg and the non-climbing leg, and establishing a double-arc reference curve under multiple constraints for the climbing leg foot-end trajectory; establishing a collision-free inequality constraint for the control point spacing based on the double-arc reference curve, setting the number of control points on the double-arc reference curve so that the control point spacing satisfies the collision-free inequality constraint for the control point spacing, and obtaining a cubic uniform B-spline curve through the control points. The cubic uniform B-spline curve is used as the swing trajectory of the climbing leg climbing the structured step terrain, and a sine curve is used as the swing trajectory of the non-climbing leg foot-end;

[0044] S203: Setting multiple obstacle crossing stages for the combined octapod robot; establishing multiple constraints on the trajectories of the front, rear, and waist joints of the first and second quadruped robots for each obstacle crossing stage based on the multiple constraints on the trajectories of the climbing leg and foot ends; obtaining the trajectories of the front, rear, and waist joints of the first and second quadruped robots for each obstacle crossing stage based on the multiple constraints on the trajectories of the front, rear, and waist joints;

[0045] S204: Complete the obstacle-crossing action according to the swinging trajectory of the climbing leg climbing the structured stepped terrain, the swinging trajectory of the foot end of the non-climbing leg, and the position trajectories of the front, rear, and waist joints of the first and second quadruped robot bodies.

[0046] Furthermore, in S201: The front part of the first quadruped robot is the front part of the octoped robot, and the rear part of the second quadruped robot is the rear part of the octoped robot; establish the foot end workspaces of the first and second quadruped robots with two-degree-of-freedom single legs, and obtain the maximum step length ranges of each leg of the first and second quadruped robots at different body pitch angles according to the quadruped robot structure and the foot end workspaces.

[0047] Furthermore, establish a three-dimensional world coordinate system. The coordinate system uses the right-hand system. The positive direction of the x-axis is the forward direction of the first quadruped robot, the positive direction of the y-axis is the left direction of the first quadruped robot, and the positive direction of the z-axis is the vertically upward direction.

[0048] Furthermore, the establishment of the foot end workspaces of the first and second quadruped robots with two-degree-of-freedom single legs, and obtaining the maximum step length ranges of each leg of the first and second quadruped robots at different body pitch angles according to the structures of the first and second quadruped robots and the simplified foot end workspaces specifically includes:

[0049] According to the lengths of the thigh and calf, the positions of the hip and knee joints, the joint angle limits, and the height h of the center of mass of the quadruped robot body s , establish the foot end workspaces of the first and second quadruped robots with two-degree-of-freedom single legs, and record the left maximum step length position coordinates of the two intersection points of the foot end workspace and the horizontal plane as (x xp , z xp ), and record the right maximum step length position coordinates as (x dp , z dp );

[0050] According to the foot end workspace, it can be obtained that the main positions restricting the foot end movement trajectory in the foot end workspace are in the upper part of the foot end workspace, and the right part of the foot end workspace has a large range and will not restrict the foot end trajectory within the set step length range. Therefore, according to the specific foot end workspaces of the first and second quadruped robots used and the task requirements, customize two rays. The endpoints of the two rays coincide and are located in the foot end workspace, and use the two rays to perform upper-edge linearization processing on the upper part of the foot end workspace to achieve linearized expression of the irregular edge of the upper part of the foot end workspace, and establish a simplified foot end workspace as Figure 2 shown. The expressions of the two rays are l r1 : z = j1x + b1 and l r2:z=k2x+b2, by solving the intersection of the two rays, the coordinates of the end points of the two rays are obtained as (x G ,z G ), where l r1 and l r2 Denote two rays, k1 and k2 represent the slopes of the two rays, and b1 and b2 represent the intercepts of the two rays with the z-axis of the coordinate system. The foot-end workspaces of the first and second quadruped robots with different body pitch angles are established. The maximum step length range of each leg group of the first and second quadruped robots when moving in the horizontal plane under the simplified foot-end workspace is obtained:

[0051] Define a waist link length as d, and the first and second quadruped robot body lengths as L;

[0052] Create a model with the fuselage pitch angle γ f =0, fuselage center of mass height h s The front legs of the quadruped robot simplify the foot end workspace, and the maximum step length range of the foot end when moving in the horizontal plane is obtained as L1, and the maximum step length position coordinate on the left is marked as (x xp1 ,z xp1 ), the coordinate of the maximum step position on the right is marked as (x dp1 ,z dp1 ), the coordinates of the endpoints of the two rays are marked as (x G1 ,z G1 ), as shown in Figure 3(a);

[0053] Create a model with the fuselage pitch angle γ f =γ, fuselage center of mass height The working space of the foot ends of the front and hind legs of the first and second quadruped robots are obtained, and the maximum step length ranges of the foot ends of the front and hind legs when moving in the horizontal plane are L2 and L3 respectively, and the maximum step length position coordinates on the left are recorded as (x xp2 ,z xp2 ) and (x xp3 ,z xp3 ), the maximum step position coordinates on the right are recorded as (x dp2 ,z dp2 ) and (x dp3 ,z dp3 ), the position coordinates of the two ray endpoints are recorded as (x G2 ,z G2 ) and (x G3 ,z G3 ), as shown in Figure 3(b) and Figure 3(c), γ represents the artificially set fuselage pitch angle. <min{M1,M3}。

[0054] Furthermore, the swing phase is divided into a climbing leg and a non-climbing leg, wherein:

[0055] The climbing leg is the leg that is currently in the state of switching from the support phase to the swing phase, performing a climbing action during the swing cycle, and the foothold is located on the structured step;

[0056] The non-climbing leg refers to the leg whose foot landing point is still on the plane where the foot end is at the beginning of the swing phase at the end of the swing phase.

[0057] Furthermore, the double arc reference curve has the starting position of the climbing leg foot end as the starting point of the double arc reference curve and the ending position of the climbing leg foot end as the ending point of the double arc reference curve;

[0058] Double arc reference curve Figure 4 As shown, the double arc reference curve consists of two arc segments with opposite bending directions. The starting point of the first arc segment is the starting point of the double arc reference curve, which is located on the horizontal ground. The end point of the second arc segment is the end point of the double arc reference curve, which is located on the structured step terrain. The intersection of the two arc segments forms the intersection point of the double arc reference curve. The intersection point is the end point of the first arc segment and the starting point of the second arc segment. The first arc segment rotates clockwise from the starting point of the first arc segment to the end point of the first arc segment, and the second arc segment rotates counterclockwise from the starting point of the second arc segment to the end point of the second arc segment. The first arc segment has O1 as the center and R1 as the radius, denoted as l1. The second arc segment has O2 as the center and R2 as the radius, denoted as l2. The centers of the two arc segments and the intersection point are collinear, and the distance between the centers of the two arcs is equal to the sum of the radii of the two arcs. The coordinates of the starting point of the climbing leg foot end (starting point of the double arc curve) are denoted as (x sp ,z sp ), the end point position (end point position of the double arc curve) is marked as (x tp ,z tp ), the coordinates of the corner position of the structured step terrain are marked as (x cp ,z cp ), the intersection point of the double arc datum curve is marked with coordinates (x jp ,z jp ), the starting angle of the double arc reference curve is recorded as The central angle of the second arc is recorded as The central angle of the first arc is recorded as The incident angle at the end point of the double arc reference curve is recorded as Define the limit position of the double arc intersection point as the intersection point of the line connecting the center point O1 of the first arc segment and the end point of the double arc reference curve with the first arc segment.

[0059] Furthermore, the climbing leg foot end trajectory has multiple constraints, including: climbing distance inequality constraint, climbing state step length inequality constraint, non-climbing state step length inequality constraint, and double arc reference curve equality and inequality constraints.

[0060] Furthermore, the climbing distance inequality constraint, the climbing state step length inequality constraint, and the non-climbing state inequality constraint refer to:

[0061] The climbing distance L d , which is the horizontal distance from the starting position of the climbing leg foot end to the structured stepped terrain, denoted as L d = x cp - x sp ;

[0062] According to the different distances from the starting position of the climbing leg foot end to the structured stepped terrain, multiple double-arc reference curves passing through the corners of the structured stepped terrain are planned. The closer the distance from the starting position of the climbing leg foot end to the structured stepped terrain, the smaller the radius of the second arc l2, the higher the double-arc intersection point in the z-axis direction, the higher the height of the curve highest point from the plane of the structured stepped terrain, and the larger the incident angle of the double-arc reference curve end point, until the double-arc intersection point moves along the first arc to the limit position of the intersection point;

[0063] To avoid the first arc l1 intersecting with the steps and being unable to plan the double-arc reference curve, the shortest horizontal distance L from the starting position of the climbing leg foot end to the structured stepped terrain is defined dmin , as shown in Figure 5 ;

[0064] Set the exit angle of the double-arc reference curve starting point as The radius of the first arc is R1. At this time, the center of the first arc can be determined by the exit angle of the starting point and the radius of the first arc, and the shortest horizontal distance L from the starting position of the climbing leg foot end to the structured stepped terrain is obtained dmin as follows:

[0065]

[0066] Set that the step lengths of each leg of the octopod robot are the same during movement, denoted by L s . The maximum step length range of each leg of the octopod robot needs to be selected according to the maximum step length ranges of each leg of the first and second quadruped robots at different fuselage pitch angles. At the same time, L2 < min{L1, L3}, so the maximum step length range of each leg of the octopod robot is selected as L2, and the maximum step length is defined as the horizontal distance L from the starting position of the foot end when switching from the support phase to the swing phase state to the left maximum step length position coordinates (x dp2 , z dp2 ), expressed as follows: smax :

[0067] L smax = L2 - (x sp - x dp2 )

[0068] The minimum step length of the climbing leg cannot be less than the shortest horizontal distance from the starting point of the climbing leg foot to the structured step terrain, which is expressed as follows:

[0069] L smin =L dmin

[0070] The step length inequality constraint for the climbing state is expressed as:

[0071] L smin ≤L s ≤L smax

[0072] The step length inequality constraint in the non-climbing state is expressed as:

[0073] 0≤L s ≤L smax

[0074] The longest horizontal distance from the starting point of the climbing leg to the structured step terrain is defined as the step length L. s , which is expressed as follows:

[0075] L dmax =L s

[0076] The climbing distance inequality constraint is expressed as:

[0077] L dmin ≤L d ≤L dmax .

[0078] Furthermore, the dual-arc reference curve equality and inequality constraints refer to:

[0079] Establish the equality constraint condition for the intersection of two arcs:

[0080]

[0081]

[0082]

[0083] After determining the starting point (starting point of the double arc reference curve), the end point (end point of the double arc reference curve), and the step length of the climbing leg, different double arc reference curves can still be planned, and the incidence angles of different end points of the double arc reference curves are also different;

[0084] Without considering the constraints of the double-ray on the double-arc reference curve, the incident angle of the end point of the double-arc reference curve is the largest when the intersection point is at the extreme position of the intersection point, which is recorded as As shown in Figure 6(a), the incident angle of the end point of the double arc reference curve is the smallest when it passes through the corner of the structured step terrain, which is recorded as As shown in Figure 6(b);

[0085] Curve equality constraint when the intersection point is at the intersection point limit position:

[0086]

[0087] in, is the coordinate of the center of the second arc;

[0088] Combined with the intersection point equality constraint, find the angle of incidence The radius R2 of the second arc and the center angles of the arcs at both ends can be used to determine the center O2 of the second arc;

[0089] The double arc reference curve passes through the structured step terrain corner (x cp ,z cp ) Curve equality constraints:

[0090]

[0091] Combined with the intersection point equality constraint, find the angle of incidence The radius R2 of the second arc and the center angles of the arcs at both ends can be used to determine the center O2 of the second arc;

[0092] When considering the constraints of the double ray on the double arc reference curve, the double arc reference curve and the ray l r2 The incident angle at the end point when it is tangent to the reverse extension line is recorded as Double arc reference curve passes through the ray endpoint (x G ,z G ) is recorded as the end point incident angle As shown in Figure 7(a) and Figure 7(b);

[0093] Double arc reference curve and ray l r2 Or curve equality constraint when its reverse extension is tangent:

[0094]

[0095] Combined with the intersection point equality constraint, for ray l r2 Or solve the double arc reference curve when its reverse extension line is tangent to it, and determine whether there is a solution:

[0096] If there is no solution, then there is no double arc reference curve that satisfies this constraint condition.

[0097] If there is a solution, then determine whether the obtained double arc reference curve satisfies the requirement that the intersection point is located on the arc from the starting position of the climbing leg foot end to the extreme position of the intersection point;

[0098] If not satisfied,

[0099] If it satisfies the requirement, the incident angle of the end point of the double arc reference curve is recorded as

[0100] At the same time, the radius R2 of the second arc and the center angles of the two arcs can be calculated, and then the position of the center O2 of the second arc can be determined;

[0101] Double arc reference curve through (x G ,z G ) when the curve equality constraint is:

[0102]

[0103] Combined with the intersection point equality constraint, the double arc reference curve passes through (x G ,z G ) to solve the double arc reference curve and determine whether there is a solution:

[0104] If there is no solution, then there is no double arc reference curve that satisfies this constraint condition.

[0105] If there is a solution, then determine whether the obtained double arc reference curve satisfies the requirement that the intersection point is located on the arc from the starting position of the climbing leg foot end to the extreme position of the intersection point;

[0106] If not satisfied,

[0107] If it satisfies the requirement, the incident angle of the end point of the double arc reference curve is recorded as

[0108] At the same time, the radius R2 of the second arc and the center angles of the two arcs can be calculated, and then the position of the center O2 of the second arc can be determined;

[0109] Define the maximum incident angle at the end point of the double arc reference curve

[0110]

[0111] Define the minimum incident angle at the end point of the double arc reference curve

[0112]

[0113] Establish the incident angle at the end point of the double arc reference curve Inequality constraints:

[0114]

[0115] The incident angle of the end point of the selected double arc reference curve is

[0116] Furthermore, the control point spacing has no collision inequality constraint, which means:

[0117]

[0118] Where ρ represents the spacing between control points on the second arc, R2 represents the radius of the second arc, and R3 represents the distance from the center of the second arc to the corner of the structured step terrain.

[0119] Establishing a collision-free inequality constraint for the control point spacing for the second arc can ensure that when the control point spacing meets the constraint, the swing trajectory of the climbing leg foot end will not collide with the structured step.

[0120] It should be understood that the double-arc reference curve has the characteristics of excessive curvature and sudden change of curvature at the intersection of the two arcs. This characteristic causes the angular velocity of the leg joints of the octapod robot to suddenly change, which is detrimental to the stability of the body.

[0121] Therefore, based on the double-arc reference curve, a cubic uniform B-spline curve is designed to overcome the problem of sudden curvature change at the intersection of the double-arc reference curve. At the same time, in order to avoid collision between the climbing leg foot end and the corner of the structured step terrain when using the B-spline curve as a swing trajectory, the position and number of control points on the second arc are designed according to the collision-free inequality constraint of the control point spacing.

[0122] The double arc reference curve is uniquely determined when the starting point and end point of the double arc reference curve, the radius of the first arc segment, the exit angle of the starting point of the double arc reference curve, and the incident angle of the end point are determined.

[0123] The present invention designs a cubic uniform B-spline curve as follows: a control point is placed at the starting point of the double arc reference curve, the midpoint of the first arc segment, the intersection point of the double arc reference curve, and the end point. The number of control points on the two arc segments is set so that the spacing between the control points on the second arc segment satisfies the control point spacing non-collision inequality constraint. The schematic diagram of the control point position is shown in FIG. Figure 8 The B-spline curve is used as the foot swing trajectory of the climbing leg, and a sine curve is designed according to the step length as the foot swing trajectory of the non-climbing leg.

[0124] The sine curve is designed as follows: Figure 9 As shown, the period of the swing phase of a single leg is set to T sw , remember t sw is the current period T sw The time in the interval is increased from 0 to T. sw , t sw ∈[0,Tsw ], the swing phase period ends t sw becomes 0 until the next swing phase cycle, t sw Then increase from 0 to T sw .

[0125] t sw The foot end position at this moment is solved as follows:

[0126]

[0127]

[0128] in, It represents the position coordinate of the foot end of the non-climbing leg during the sinusoidal swing process, H sw Indicates the swing height of the non-climbing leg foot end, L s Indicates the step size.

[0129] Furthermore, the octapod robot is set to adopt a static gait when climbing structured step terrain, and the static gait phase switching relationship of the robot is established, that is, at most only one leg is in the swing phase at the same time, and the other legs are in the support phase.

[0130] Multiple obstacle crossing stages are set for the combined octapod robot, and the multiple obstacle crossing stages include:

[0131] Phase 1: The first quadruped robot lifting phase;

[0132] Phase 2: The first quadruped robot front legs climb the stage;

[0133] Phase 3: The first quadruped robot's hind legs climb the stage;

[0134] Phase 4: The first quadruped robot resumes its translational motion, while the second quadruped robot is elevated;

[0135] Stage 5: The second quadruped robot’s front legs climb the stage;

[0136] Stage 6: The second quadruped robot's hind legs climb the stage;

[0137] Phase 7: The second quadruped robot resumes translational motion.

[0138] The robot bodies are numbered as shown in Figure 10(a). A, B, C, D, and E all represent body position parameters. Point A represents the position of the center point of the front end of the first quadruped robot body, point B represents the position of the center point of the rear end of the first quadruped robot body, point C represents the position of the center point of the waist pitch joint, point D represents the position of the center point of the front end of the second quadruped robot body, and point E represents the position of the center point of the rear end of the second quadruped robot body. Point F is on the outside of the second robot body, and the distance between point F and point E is d. Points F, D, and E are on the same straight line. (x A ,z A )、(x B ,z B )、(x C ,z C )、(x D ,z D )、(x E ,z E )、(x f ,z F ) represent the position coordinates of points A, B, C, D, E, and F respectively.

[0139] The process of an octapod robot climbing structured stepped terrain is divided into seven stages. Multi-equation constraints on the position trajectory of the robot are established for each stage, and obstacle climbing action planning is performed to achieve obstacle climbing planning under multiple constraints.

[0140] Furthermore, multiple constraints are established for the position trajectories of the front, rear, and waist joints of the first and second quadruped robots for each obstacle crossing stage, wherein the multiple constraints in stage one are:

[0141] x A =x+(L+d)cos(tΔγ)

[0142] z A =z+(L+d)sin(tΔγ)

[0143] x B =x+dcos(tΔγ)

[0144] z B =z+dsin(tΔγ)

[0145] z=h d

[0146] x D =xd

[0147] z D =h d

[0148] c E =xLd

[0149] z E =h d

[0150] The position trajectory of the fuselage in this stage can be obtained by the multiple constraints of the fuselage position parameter trajectory.

[0151] Phase 1: Belongs to the posture adjustment phase, set the step length to L s , the fuselage length is L, (x, z) represents the coordinates of the waist pitch joint position, and in stage 1, the left and right front legs of the first quadruped robot both meet the climbing distance inequality constraints and the climbing state step length inequality constraints; the waist pitch joint angle during the climbing process is set to be γ, and the waist pitch joint angle during the translation process is set to be 0; in this stage, the pitch angle of the first quadruped robot body is adjusted from 0 during the translation process to γ f =γ, the pitch angle of the second quadruped robot body remains at 0;

[0152] First, keep the position of the waist pitch joint C unchanged, and rotate the first quadruped robot clockwise around the waist pitch joint C until the pitch angle of the first quadruped robot reaches the set γ f The second quadruped robot keeps the height h of the body during the translation phase. d During the rotation, the trajectory of point A is an arc with C as the center and AC as the radius, and the trajectory of point B is an arc with C as the center and BC as the radius. Let T be a complete cycle of the swing phase and the support phase of a single leg, and t be the moment in the current cycle T. Starting from 0, the unit time is increased each time, and finally increases to T, t∈[0,T]. After increasing to T, t becomes 0, and t continues to increase from 0 to T at the beginning of the next complete cycle. The rotation angle of the waist pitch joint per unit time is recorded as

[0153] Furthermore, for each obstacle crossing stage, multiple constraints are established for the position trajectories of the front, rear, and waist joints of the first and second quadruped robots. In particular, stage two is the climbing stage of the front legs of the first quadruped robot, and stage two is the movement stage. In stage two, the position parameters of the eight-legged robot body remain unchanged in the z direction, and the position changes per unit time along the x direction. Keeping the height of the first quadruped robot body unchanged, set the left front leg and the right front leg of the first quadruped robot as climbing legs in turn, and use the climbing leg foot end swing trajectory to perform the climbing action until the left front leg and the right front leg of the first quadruped robot both climb onto the structured step terrain.

[0154] Furthermore, for each obstacle crossing stage, multiple constraints are established for the position trajectories of the front, rear, and waist joints of the first and second quadruped robots. In particular, stage three is the climbing stage of the hind legs of the first quadruped robot, and stage three belongs to the movement stage. In stage three, the position parameters of the octapod robot body remain unchanged in the z direction, and the position changes per unit time along the x direction. Using a static gait and non-climbing leg-foot end swing trajectory, and adjusting the leg-foot end positions of the octapod robot according to the non-climbing step-length inequality constraints, the left and right hind legs of the first quadruped robot ultimately satisfy the climbing distance inequality constraints and the climbing step-length inequality constraints.

[0155] The left hind leg and the right hind leg of the first quadruped robot are set as climbing legs in turn, the height of the first quadruped robot body is kept unchanged, and the climbing leg foot end adopts the climbing leg foot end swing trajectory to perform the climbing action until the left hind leg and the right hind leg of the first quadruped robot both climb onto the structured step terrain.

[0156] Furthermore, multiple constraints are established for the position trajectories of the front, rear, and waist joints of the first and second quadruped robots for each obstacle crossing stage, wherein the constraints for stage four include:

[0157] x A =x+(L+d)cos(γ-tΔγ)

[0158] z A =z+(L+d)sin(γ-tΔγ)

[0159] x B =x+dcos(γ-tΔγ)

[0160] z B =z+dsin(γ-tΔγ)

[0161] x=x F +(L+2d)cos(tΔγ)

[0162] z=z F +(L+2d)sin(tΔγ)

[0163] x D =x F +(L+d)cos(tΔγ)

[0164] z D =z F +(L+d)sin(tΔγ)

[0165] x E =x F +dcos(tΔγ)

[0166] zE =z F +dsin(tΔγ)

[0167] z F =h d

[0168] The position trajectory of the fuselage in this stage can be obtained by the multiple constraints of the fuselage position parameter trajectory.

[0169] Phase 4: The first quadruped robot resumes its translational motion, and the second quadruped robot is raised, as shown in Figure 10(b). Phase 4 is the posture adjustment phase. The second quadruped robot performs the same actions as the first quadruped robot. In this phase, the first quadruped robot's body pitch angle is adjusted from γ to γ. f =γ is adjusted to 0, and the pitch angle of the second quadruped robot is adjusted from 0 to γ f =γ. The second quadruped robot rotates clockwise around point F. Therefore, the motion trajectory of point C is an arc with point F as the center and radius CF as the radius. The motion trajectory of point D is an arc with point F as the center and radius DF as the radius. The motion trajectory of point E is an arc with point F as the center and radius EF as the radius. Because the angles of the waist links of the first and second quadruped robots with the horizontal plane change in opposite directions during adjustment, the positions of points A and B can both be represented by the position of point C.

[0170] Furthermore, for each obstacle crossing stage, multiple constraints are established for the position trajectories of the front, rear and waist joints of the first and second quadruped robots, wherein stage five is the climbing stage of the front legs of the second quadruped robot; stage five belongs to the movement stage; in stage five, the position parameters of the eight-legged robot body remain unchanged in the z direction, and the position unit time change along the x direction is Using a static gait and a non-climbing leg-foot swing trajectory, the octapod's leg-foot positions are adjusted according to the non-climbing step-length inequality constraints, ensuring that both the left and right front legs of the second quadruped robot ultimately meet the climbing distance inequality constraints and the climbing step-length inequality constraints. The left and right front legs of the second quadruped robot are sequentially set as climbing legs, maintaining the second quadruped's body height unchanged. The climbing leg-foot swing trajectory is used to execute the climbing maneuver until both the left and right front legs of the second quadruped robot have reached the structured step terrain.

[0171] Furthermore, for each obstacle crossing stage, multiple constraints are established for the position trajectories of the front, rear, and waist joints of the first and second quadruped robots. In stage six, the second quadruped robot's hind legs climb, and stage six belongs to the movement stage. In stage six, the position parameters of the eight-legged robot body remain unchanged in the z direction, and the position changes per unit time along the x direction. A static gait and a non-climbing leg-foot-end swinging trajectory are adopted, and the positions of the leg-foot ends of the octapod robot are adjusted according to the non-climbing state step length inequality constraints, so that the left hind leg and the right hind leg of the second quadruped robot finally meet the climbing distance inequality constraints and the climbing state step length inequality constraints; the left hind leg and the right hind leg of the second quadruped robot are set as climbing legs in turn, and the height of the second quadruped robot body is kept unchanged. The climbing leg-foot-end swinging trajectory is adopted to perform the climbing action until the left hind leg and the right hind leg of the second quadruped robot both climb onto the structured step terrain.

[0172] Furthermore, multiple constraints are established for the position trajectories of the front, rear, and waist joints of the first and second quadruped robots for each obstacle crossing stage, wherein the constraints of stage seven include:

[0173] z A =z B =z=h d +H

[0174] x D =x-dcos(γ-tΔγ)

[0175] z D =z-dsin(γ-tΔγ)

[0176] x E =x-(L+d)cos(γ-tΔγ)

[0177] z E =z-(L+d)sin(γ-tΔγ)

[0178] The position trajectory of the fuselage in this stage can be obtained by the multiple constraints of the fuselage position parameter trajectory.

[0179] Where H represents the height of the structured step terrain.

[0180] Phase 7: The second quadruped robot resumes translation. As shown in Figure 10(c), Phase 7 belongs to the posture adjustment phase. In Phase 7, the pitch angle of the first quadruped robot is kept at 0, and the pitch angle of the second quadruped robot is adjusted from γ f =γ is adjusted to 0. First, keep the position of point C unchanged, and the second quadruped robot rotates counterclockwise around point C until the pitch angle of the second quadruped robot body is 0, and the first quadruped robot maintains the height h of the translation stage. d The trajectory of point D is an arc with center C and radius CD, and the trajectory of point E is an arc with center C and radius CE.

[0181] Figure 10(a)-Figure 10(c)A′, A″, B′, B″, C′, C″, D′, D″, E′, E″ represent the positions of the octapod robot’s A, B, C, D, E, and F during the change process at the current stage.

[0182] The above is the multi-constraint motion planning method for the combined octapode robot with waist pitch freedom, which realizes the coordinated control of the combined octapode robot in the process of obstacle crossing, so that the octapode robot has good terrain adaptability and coordinated control capabilities.

[0183] The motion planning of the present invention includes two parts: swing phase foot end trajectory planning and obstacle crossing planning. Obstacle crossing planning includes fuselage position trajectory planning and obstacle crossing action planning, which specifically includes the following two aspects:

[0184] The foot-end trajectory planning of the swing phase is performed under multiple constraints. Structured step terrain is used as the obstacle terrain. First, the swing phase is divided into the climbing leg and the non-climbing leg, and each is defined separately. Second, a double-arc reference curve is established under multiple constraints, including inequality constraints on the climbing distance, step size, and the double-arc reference curve. Finally, based on the double-arc reference curve, a collision-free inequality constraint on the control point spacing is established. The number of control points on the double-arc reference curve is set so that the control point spacing satisfies the collision-free inequality constraint. A cubic uniform B-spline curve is then derived from the control points and used as the swing trajectory of the climbing leg climbing the structured step terrain. A sine curve is used as the foot-end swing trajectory of the non-climbing leg.

[0185] Obstacle traversal planning under multiple constraints. This paper sets the gait type of a modular octapod robot when climbing structured stepped terrain. The obstacle traversal process is divided into a movement phase and a posture adjustment phase, which are further subdivided into seven sub-phases. Multiple constraints are designed in each phase to perform body position trajectory planning and obstacle traversal motion planning.

[0186] The motion planning of the eight-legged robot during obstacle crossing is divided into two parts: foot-end trajectory planning for the swing phase and obstacle crossing planning. In the foot-end trajectory planning for the swing phase, by dividing the swing phase into climbing legs and non-climbing legs, a double-arc reference curve suitable for multi-constraint situations is constructed, and then a cubic uniform B-spline curve is designed as the foot-end trajectory of the climbing leg. The number of control points is adjusted according to the constraints to better adapt to the obstacle terrain; in the obstacle crossing planning, by dividing the obstacle crossing process into the motion phase and the posture adjustment phase, multiple constraints are designed in stages, and the body position trajectory planning and obstacle crossing action planning are carried out to improve the ability to pass through the obstacle terrain. The multi-constraint motion planning method established by the present invention fully considers the impact of the obstacle terrain on the foot-end trajectory and the body posture during the obstacle crossing process, thereby improving the terrain adaptability of the combined eight-legged robot.

[0187] Example 2: This embodiment provides a multi-constraint motion planning system for a combined octapod robot;

[0188] Combined octapod robot multi-constraint motion planning system, including:

[0189] A setting module is configured to: divide the combined octapod robot into two front and rear quadruped robots: a first quadruped robot and a second quadruped robot;

[0190] A curve establishment module is configured to: take the structured step terrain as the obstacle terrain, divide the swing phase into the climbing leg and the non-climbing leg, establish a double-arc reference curve under multiple constraints of the climbing leg foot end trajectory, establish a collision-free inequality constraint on the control point spacing based on the double-arc reference curve, set the number of control points on the double-arc reference curve so that the control point spacing satisfies the collision-free inequality constraint on the control point spacing, obtain a cubic uniform B-spline curve through the control points, use the cubic uniform B-spline curve as the swing trajectory of the climbing leg climbing the structured step terrain, and use the sine curve as the swing trajectory of the non-climbing leg foot end;

[0191] The trajectory generation module is configured to: establish multiple constraints on the position trajectories of the front, rear, and waist joints of the first and second quadruped robots for each obstacle crossing stage based on the multiple constraints on the climbing leg and foot trajectories; obtain the position trajectories of the front, rear, and waist joints of the first and second quadruped robots for each obstacle crossing stage based on the multiple constraints on the position trajectories of the front, rear, and waist joints;

[0192] The obstacle crossing module is configured to complete the obstacle crossing action according to the swing trajectory of the climbing legs climbing the structured step terrain, the swing trajectory of the non-climbing legs and the front, rear and waist joint position trajectories of the front and rear quadruped robot bodies.

[0193] Embodiment 3: This embodiment provides a combined octapod robot;

[0194] The combined octapod robot includes a processor and a memory. The processor includes, but is not limited to, any one of the following: a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), and an FPGA (Field Programmable Gate Array). The memory may include storage devices such as RAM (Random Access Memory) and ROM (Read Only Memory). The processor and the memory may be connected via a system bus.

[0195] In an exemplary embodiment, the memory stores at least one instruction, at least one program, code set or instruction set, and the at least one instruction, the at least one program, the code set or instruction set is loaded and executed by the processor to implement the above-mentioned robot motion control method.

[0196] The octapod robot comprises: a first quadruped robot and a second quadruped robot connected to each other;

[0197] Taking structured step terrain as obstacle terrain, the legs of the octapod robot as the swing phase are divided into climbing legs and non-climbing legs, and a double-arc reference curve under multiple constraints of the climbing leg foot-end trajectory is established. The multiple constraints of the climbing leg foot-end trajectory include: climbing distance inequality constraint, climbing state step length inequality constraint, non-climbing state step length inequality constraint, and double-arc reference curve inequality constraint;

[0198] Based on the double-arc reference curve, a collision-free inequality constraint for the distance between control points is established, and then a cubic uniform B-spline curve is obtained. The cubic uniform B-spline curve is used as the swing trajectory of the climbing leg climbing the structured step terrain, and the sine curve is used as the swing trajectory of the foot end of the non-climbing leg.

[0199] Based on the multi-constraint conditions of the climbing leg foot end trajectory, the multi-constraint conditions of the position trajectory of the front, rear and waist joints of the first and second quadruped robots are established for each obstacle crossing stage; based on the multi-constraint conditions of the position trajectory of the front, rear and waist joints of the first and second quadruped robots, the position trajectory of the front, rear and waist joints of the first and second quadruped robots are obtained for each obstacle crossing stage;

[0200] The obstacle crossing action is completed according to the swing trajectory of the climbing legs climbing the structured step terrain, the swing trajectory of the non-climbing legs and the position trajectory of the front, rear and waist joints of the front and rear quadruped robot body.

[0201] Furthermore, the first quadruped robot is equipped with four legs, namely: a first left front leg, a first right front leg, a first left hind leg and a first right hind leg;

[0202] The second quadruped robot is equipped with four legs, namely: a second left front leg, a second right front leg, a second left hind leg and a second right hind leg;

[0203] Each leg consists of a hip joint, thigh joint, knee joint, and calf joint connected in sequence;

[0204] The center point of the line connecting the left rear hip joint and the right rear hip joint is fixedly connected to the first end of the first vertical rod;

[0205] The center point of the line connecting the left front hip joint and the right front hip joint is fixedly connected to the first end of the second vertical rod;

[0206] The second end of the first vertical rod is hinged to the second end of the second vertical rod, and the hinged portion has pitch freedom, and the hinged portion is a waist pitch joint.

[0207] Furthermore, the hip joints and knee joints are each provided with a drive controller; the four hip joint drive controllers and the four knee joint drive controllers of the first quadruped robot are each connected to the first quadruped robot controller;

[0208] The four hip joint drive controllers and the four knee joint drive controllers of the second quadruped robot are all connected to the second quadruped robot controller;

[0209] The first quadruped robot controller and the second quadruped robot controller are both connected to a host computer.

[0210] Schematic diagram of a combined octapod robot with waist pitch freedom, as shown in Figure 11 As shown, the top view is Figure 12 As shown, it is composed of two quadruped robots connected in series front and back. The series connection mechanism adopts two rigid connecting rods. One end of the rigid connecting rod of the first quadruped robot is connected to the midpoint of the line connecting the two hip joints of the rear legs of the fuselage, and one end of the rigid connecting rod of the second quadruped robot is connected to the midpoint of the line connecting the two hip joints of the front legs of the fuselage. The connection between the connecting rod and the fuselage is a rigid connection without degree of freedom. The other ends of the two rigid connecting rods are connected to each other to form a waist pitch joint, which has one pitch degree of freedom.

[0211] The eight legs of the combined octapod robot are divided into four groups, namely the front legs of the first quadruped robot, the hind legs of the first quadruped robot, the front legs of the second quadruped robot, and the hind legs of the second quadruped robot. Each group includes one leg on each of the left and right sides of the body.

[0212] In an exemplary embodiment, a computer-readable storage medium is also provided, in which at least one instruction, at least one program, a code set or an instruction set is stored. When the at least one instruction, the at least one program, the code set or the instruction set is executed by a processor of a computer device, the above-mentioned robot motion control method is implemented.

[0213] Optionally, the computer-readable storage medium may include: ROM (Read Only Memory), RAM (Random Access Memory), SSD (Solid State Drives), or an optical disk, etc. Among them, the random access memory may include ReRAM (Resistance Random Access Memory) and DRAM (Dynamic Random Access Memory).

[0214] In an exemplary embodiment, a computer program product or computer program is also provided, the computer program product or computer program including computer instructions stored in a computer-readable storage medium. A processor of a robot reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the robot to perform the above-described robot motion control method.

[0215] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art will readily appreciate that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention are intended to be within the scope of protection of the present invention.

Claims

1. A multi-constraint motion planning method for a combined octapod robot, characterized by: include: The combined octapod robot is divided into two front and rear quadruped robots: a first quadruped robot and a second quadruped robot; The structured step terrain is taken as the obstacle terrain, the swing phase is divided into the climbing leg and the non-climbing leg, and a double arc reference curve is established under the multi-constraint conditions of the climbing leg foot end trajectory. On the basis of the double arc reference curve, a collision-free inequality constraint for the control point spacing is established. The number of control points on the double arc reference curve is set so that the control point spacing satisfies the collision-free inequality constraint for the control point spacing. A cubic uniform B-spline curve is obtained through the control points. The cubic uniform B-spline curve is used as the swing trajectory of the climbing leg climbing the structured step terrain, and the sine curve is used as the swing trajectory of the non-climbing leg foot end. Set up multiple obstacle crossing stages for the combined octapod robot; establish multiple constraints on the position trajectories of the front, rear, and waist joints of the first and second quadruped robots for each obstacle crossing stage based on the multiple constraints on the trajectories of the climbing leg ends; Based on the multi-constraint conditions of the front, rear and waist joint position trajectories of the quadruped robot's body, the front, rear and waist joint position trajectories of the quadruped robot are obtained before and after each obstacle crossing stage. The climbing leg foot trajectory has multiple constraints, including: climbing distance inequality constraint, climbing state step length inequality constraint, non-climbing state step length inequality constraint; The step length inequality constraint for the climbing state is expressed as: in, represents the step length of each leg during the movement of the octapod robot, Indicates the minimum step size; Minimum step size as follows: in, It represents the shortest horizontal distance from the starting point of the climbing leg to the structured step terrain; as follows: in, is the radius of the first arc, is the starting point angle of the double arc reference curve; Maximum step length Defined as the coordinate from the starting point of the foot to the maximum step length on the left when switching from the stance phase to the swing phase The horizontal distance is expressed as follows: in, Indicates the maximum step length range of each leg of the octapod robot, Indicates that the starting point of the climbing leg is Position coordinates in direction; The step length inequality constraint in the non-climbing state is expressed as: The longest horizontal distance from the starting point of the climbing leg to the structured step terrain Defined as step size , which is expressed as follows: Climbing distance The inequality constraints are expressed as: ； Set up multiple obstacle crossing stages for the combined octapod robot; for each obstacle crossing stage, establish multiple constraint conditions for the position trajectories of the front, rear and waist joints of the front and rear quadruped robots, point A represents the initial position point of the center point of the front end of the first quadruped robot body, point B represents the initial position point of the center point of the rear end of the first quadruped robot body, point C represents the initial position point of the center point of the waist pitch joint, point D represents the initial position point of the center point of the front end of the second quadruped robot body, point E represents the initial position point of the center point of the rear end of the second quadruped robot body, point F is on the outside of the second quadruped robot body, and the distance between point F and point E is , points F, D and E are on the same straight line, 、 、 、 、 、 Represent the position coordinates of points A, B, C, D, E, and F respectively; among them, the multi-constraint conditions in stage one are: The position trajectory of the fuselage in this stage can be obtained by the multiple constraints of the fuselage position parameter trajectory; Phase 1 is the posture adjustment phase, and the step length is set to , the fuselage length is , Represents the position coordinates of the waist pitch joint. Assume that in stage 1, the left and right front legs of the first quadruped robot both meet the climbing distance inequality constraint and the climbing state step length inequality constraint; set the waist pitch joint angle during the climbing process to be , the waist pitch joint angle during translation remains at 0, Indicates the manually set pitch angle of the fuselage; in this stage, the pitch angle of the first quadruped robot is adjusted from 0 in the translation process to , the pitch angle of the second quadruped robot body remains at 0; First, keep the position of the waist pitch joint C unchanged, and rotate the first quadruped robot clockwise around the waist pitch joint C until the first quadruped robot body pitch angle reaches the set The second quadruped robot keeps the height of the body during the translation phase. In the process of rotation, the trajectory of point A is an arc with C as the center and AC as the radius, and the trajectory of point B is an arc with C as the center and BC as the radius. Assume that a complete cycle of the swing phase and the support phase of a single leg is ,remember For the current cycle The moment in time, starting from 0 and increasing by unit time each time, eventually increases to , , increases to back Becomes 0 and the next full cycle begins Continue to increase from 0 to ; The rotation angle of the waist pitch joint per unit time is recorded as ; The obstacle crossing action is completed according to the swing trajectory of the climbing legs climbing the structured step terrain, the swing trajectory of the non-climbing legs and the position trajectory of the front, rear and waist joints of the front and rear quadruped robot body.

2. The multi-constraint motion planning method for a combined octapod robot according to claim 1, wherein: The climbing leg foot end trajectory has multiple constraints, including: double arc reference curve equality and inequality constraints; Establish the equality constraint condition for the intersection of two arcs: The first arc is As the radius, the second arc is The radius of the two arcs is φ, and the centers of the two arcs and the intersection point are collinear, and the distance between the centers of the two arcs is equal to the sum of the radii of the two arcs; the coordinates of the starting point of the climbing leg foot end are marked as , the coordinates of the end point are marked as The starting angle of the double arc reference curve is recorded as , the central angle of the second arc is recorded as , the central angle of the first arc is recorded as The incident angle at the end point of the double arc reference curve is recorded as , define the limit position of the intersection point of the double arcs as the center of the first arc The intersection of the line connecting the end point of the double arc datum curve and the first arc segment; Curve equality constraint when the intersection point is at the intersection point limit position: in, is the coordinate of the center of the second arc; Combined with the intersection point equality constraint, the incident angle of the end point of the double arc reference curve is calculated ; Double arc datum curve passing through the corner of structured step terrain When the curve equality constraint: in, Indicates the coordinates of the corner position of the structured step terrain; Combined with the intersection point equality constraint, the incident angle of the end point of the double arc reference curve is calculated ; Double arc datum curve and ray Or curve equality constraint when its reverse extension is tangent: in, Represents rays The slope of Represents rays and coordinate system The intercept of the axis; Combined with the intersection point equality constraint, the incident angle of the end point of the double arc reference curve is calculated ; Double arc datum curve passes through double ray endpoints Curve equality constraints when : Combined with the intersection point equality constraint, the incident angle of the end point of the double arc reference curve is calculated ; Define the maximum incident angle at the end point of the double arc reference curve : in, Indicates the incident angle of the end point of the double arc reference curve when the intersection point is at the extreme position of the intersection point. Represents double arc datum curve and ray Or the incident angle at the end point when its reverse extension line is tangent, Indicates that the double arc datum curve passes through the ray endpoint The end point incident angle at ; Define the minimum incident angle at the end point of the double arc reference curve : in, Indicates the incident angle of the end point of the double arc reference curve when it passes through the corner of the structured step terrain; Establish the incident angle at the end point of the double arc reference curve Inequality constraints: The incident angle of the end point of the selected double arc reference curve is ; The control point spacing non-collision inequality constraint means: in, Indicates the spacing between control points on the second arc segment. Indicates the radius of the second arc, Indicates the distance from the center of the second arc segment to the corner of the structured stepped terrain.

3. The multi-constraint motion planning method for a combined octapod robot according to claim 1, wherein: The multiple constraints of the front, rear and waist joint position trajectories of the first and second quadruped robots are established for each obstacle crossing stage, wherein the second stage is the climbing stage of the front legs of the first quadruped robot, which belongs to the movement stage; in the second stage, the position parameters of the octapod robot body are The direction remains unchanged, along Change in position per unit time of direction , is the period of the swing phase; keeping the height of the first quadruped robot body unchanged, sequentially setting the left front leg and the right front leg of the first quadruped robot as climbing legs, and using the climbing leg foot end swing trajectory to perform the climbing action until the left front leg and the right front leg of the first quadruped robot both climb onto the structured step terrain; The multiple constraints of the front, rear and waist joint position trajectories of the first and second quadruped robots are established for each obstacle crossing stage, wherein the third stage is the climbing stage of the hind legs of the first quadruped robot. The third stage belongs to the movement stage. The position parameters of the octapod robot body are The direction remains unchanged, along Change in position per unit time of direction Adopting a static gait and a non-climbing leg-foot end swing trajectory, and adjusting the leg-foot end positions of the octapod robot according to the non-climbing state step length inequality constraint, the left and right hind legs of the first quadruped robot ultimately satisfy the climbing distance inequality constraint and the climbing state step length inequality constraint; The left hind leg and the right hind leg of the first quadruped robot are set as climbing legs in turn, the height of the first quadruped robot body is kept unchanged, and the climbing leg foot end adopts the climbing leg foot end swing trajectory to perform the climbing action until the left hind leg and the right hind leg of the first quadruped robot both climb onto the structured step terrain.

4. The multi-constraint motion planning method for a combined octapod robot according to claim 1, wherein: The multiple constraints for the position trajectories of the front, rear, and waist joints of the first and second quadruped robots are established for each obstacle crossing stage, wherein the constraints for stage four include: The position trajectory of the fuselage in this stage can be obtained by the multiple constraints of the fuselage position parameter trajectory; Phase 4: The first quadruped robot resumes its horizontal motion, and the second quadruped robot is raised. Phase 4 is the posture adjustment phase. The second quadruped robot performs the same actions as the first quadruped robot. In this phase, the first quadruped robot's body pitch angle is adjusted from Adjust to 0, the pitch angle of the second quadruped robot body is adjusted from 0 to ; The second quadruped robot rotates clockwise around point F, so the motion trajectory of point C is an arc with point F as the center and CF as the radius, the motion trajectory of point D is an arc with point F as the center and DF as the radius, and the motion trajectory of point E is an arc with point F as the center and EF as the radius.

5. The multi-constraint motion planning method for a combined octapod robot according to claim 1, wherein: The multi-constraint conditions for the position trajectory of the front, rear and waist joints of the first and second quadruped robots are established for each obstacle crossing stage, wherein the fifth stage is the climbing stage of the front legs of the second quadruped robot; the fifth stage belongs to the movement stage; the position parameters of the eight-legged robot body are The direction remains unchanged, along Change in position per unit time of direction , is the period of the swing phase; a static gait and a non-climbing leg foot end swing trajectory are adopted, and the position of each leg foot end of the octapod robot is adjusted according to the non-climbing state step length inequality constraint, so that the left front leg and the right front leg of the second quadruped robot finally satisfy the climbing distance inequality constraint and the climbing state step length inequality constraint; the left front leg and the right front leg of the second quadruped robot are set as climbing legs in turn, and the height of the second quadruped robot body is kept unchanged. The climbing leg foot end adopts the climbing leg foot end swing trajectory to perform the climbing action until the left front leg and the right front leg of the second quadruped robot both climb onto the structured step terrain; The multi-constraint conditions for the front, rear and waist joint position trajectories of the front and rear quadruped robot bodies are established for each obstacle crossing stage, wherein the sixth stage is the second quadruped robot hind leg climbing stage, the sixth stage belongs to the movement stage, and the eight-legged robot body position parameters are The direction remains unchanged, along Change in position per unit time of direction ; A static gait and a non-climbing leg foot end swing trajectory are adopted, and the position of each leg foot end of the eight-legged robot is adjusted according to the non-climbing state step length inequality constraint, so that the left hind leg and the right hind leg of the second quadruped robot finally meet the climbing distance inequality constraint and the climbing state step length inequality constraint; the left hind leg and the right hind leg of the second quadruped robot are set as climbing legs in turn, and the height of the second quadruped robot body is kept unchanged. The climbing leg foot end adopts the climbing leg foot end swing trajectory to perform the climbing action until the left hind leg and the right hind leg of the second quadruped robot both climb onto the structured step terrain; The multiple constraints for the position trajectories of the front, rear, and waist joints of the front and rear quadruped robots are established for each obstacle crossing stage, wherein the constraints for stage seven include: The position trajectory of the fuselage in this stage can be obtained by the multiple constraints of the fuselage position parameter trajectory; in, Indicates the height of structured step terrain; Phase 7: The second quadruped robot resumes its translation phase. Phase 7 is the posture adjustment phase. In Phase 7, the first quadruped robot body pitch angle is kept at 0, and the second quadruped robot body pitch angle is changed from Adjust to 0; first keep the position of point C unchanged, the second quadruped robot rotates counterclockwise around point C until the pitch angle of the second quadruped robot body is 0, and the first quadruped robot maintains the height of the translation stage unchanged; the trajectory of point D is an arc with C as the center and CD as the radius, and the trajectory of point E is an arc with C as the center and CE as the radius.

6. The robot multi-constraint motion planning system is characterized by: The system comprises: A setting module is configured to: divide the combined octapod robot into two front and rear quadruped robots: a first quadruped robot and a second quadruped robot; A curve establishment module is configured to: take the structured step terrain as the obstacle terrain, divide the swing phase into the climbing leg and the non-climbing leg, establish a double-arc reference curve under multiple constraints of the climbing leg foot end trajectory, establish a collision-free inequality constraint on the control point spacing based on the double-arc reference curve, set the number of control points on the double-arc reference curve so that the control point spacing satisfies the collision-free inequality constraint on the control point spacing, obtain a cubic uniform B-spline curve through the control points, use the cubic uniform B-spline curve as the swing trajectory of the climbing leg climbing the structured step terrain, and use the sine curve as the swing trajectory of the non-climbing leg foot end; The climbing leg foot trajectory has multiple constraints, including: climbing distance inequality constraint, climbing state step length inequality constraint, non-climbing state step length inequality constraint; The step length inequality constraint for the climbing state is expressed as: in, represents the step length of each leg during the movement of the octapod robot, Indicates the minimum step size; Minimum step size as follows: in, It represents the shortest horizontal distance from the starting point of the climbing leg to the structured step terrain; as follows: in, is the radius of the first arc, is the starting point angle of the double arc reference curve; Maximum step length Defined as the coordinate from the starting point of the foot to the maximum step length on the left when switching from the stance phase to the swing phase The horizontal distance is expressed as follows: in, Indicates the maximum step length range of each leg of the octapod robot, Indicates that the starting point of the climbing leg is Position coordinates in direction; The step length inequality constraint in the non-climbing state is expressed as: The longest horizontal distance from the starting point of the climbing leg to the structured step terrain Defined as step size , which is expressed as follows: Climbing distance The inequality constraints are expressed as: ； Set up multiple obstacle crossing stages for the combined octapod robot; for each obstacle crossing stage, establish multiple constraint conditions for the position trajectories of the front, rear and waist joints of the front and rear quadruped robots, point A represents the initial position point of the center point of the front end of the first quadruped robot body, point B represents the initial position point of the center point of the rear end of the first quadruped robot body, point C represents the initial position point of the center point of the waist pitch joint, point D represents the initial position point of the center point of the front end of the second quadruped robot body, point E represents the initial position point of the center point of the rear end of the second quadruped robot body, point F is on the outside of the second quadruped robot body, and the distance between point F and point E is , points F, D and E are on the same straight line, 、 、 、 、 、 Represent the position coordinates of points A, B, C, D, E, and F respectively; among them, the multi-constraint conditions in stage one are: The position trajectory of the fuselage in this stage can be obtained by the multiple constraints of the fuselage position parameter trajectory; Phase 1 is the posture adjustment phase, and the step length is set to , the fuselage length is , Represents the position coordinates of the waist pitch joint. Assume that in stage 1, the left and right front legs of the first quadruped robot both meet the climbing distance inequality constraint and the climbing state step length inequality constraint; set the waist pitch joint angle during the climbing process to be , the waist pitch joint angle during translation remains at 0, Indicates the manually set pitch angle of the fuselage; in this stage, the pitch angle of the first quadruped robot is adjusted from 0 in the translation process to , the pitch angle of the second quadruped robot body remains at 0; First, keep the position of the waist pitch joint C unchanged, and rotate the first quadruped robot clockwise around the waist pitch joint C until the first quadruped robot body pitch angle reaches the set The second quadruped robot keeps the height of the body during the translation phase. In the process of rotation, the trajectory of point A is an arc with C as the center and AC as the radius, and the trajectory of point B is an arc with C as the center and BC as the radius. Assume that a complete cycle of the swing phase and the support phase of a single leg is ,remember For the current cycle The moment in time, starting from 0 and increasing by unit time each time, eventually increases to , , increases to back Becomes 0 and the next full cycle begins Continue to increase from 0 to ; The rotation angle of the waist pitch joint per unit time is recorded as ; The trajectory generation module is configured to: set multiple obstacle crossing stages for the combined octapod robot; establish multiple constraints for the position trajectories of the front, rear, and waist joints of the first and second quadruped robots for each obstacle crossing stage based on the multiple constraints for the trajectories of the climbing leg and foot ends; and obtain the position trajectories of the front, rear, and waist joints of the first and second quadruped robots for each obstacle crossing stage based on the multiple constraints for the position trajectories of the front, rear, and waist joints of the first and second quadruped robots. The obstacle crossing module is configured to complete the obstacle crossing action according to the swing trajectory of the climbing leg climbing the structured step terrain, the swing trajectory of the foot end of the non-climbing leg and the position trajectory of the front, rear and waist joints of the first and second quadruped robots.

7. The robot multi-constraint motion planning system according to claim 6, characterized in that: The climbing leg foot end trajectory has multiple constraints, including: double arc reference curve equality and inequality constraints; Establish the equality constraint condition for the intersection of two arcs: The first arc is As the radius, the second arc is The radius of the two arcs is θ, and the centers of the two arcs and the intersection point are collinear, and the distance between the centers of the two arcs is equal to the sum of the radii of the two arcs. The coordinates of the starting point of the climbing leg are marked as , the coordinates of the end point are marked as The starting angle of the double arc reference curve is recorded as , the central angle of the second arc is recorded as , the central angle of the first arc is recorded as The incident angle at the end point of the double arc reference curve is recorded as , define the limit position of the double arc intersection point as the center of the first arc The intersection of the line connecting the end point of the double arc datum curve and the first arc segment; Curve equality constraint when the intersection point is at the intersection point limit position: in, is the coordinate of the center of the second arc; Combined with the intersection point equality constraint, the incident angle of the end point of the double arc reference curve is calculated ; Double arc datum curve passing through structured step terrain corner When the curve equality constraint: in, Indicates the coordinates of the corner position of the structured step terrain; Combined with the intersection point equality constraint, the incident angle of the end point of the double arc reference curve is calculated ; Double arc datum curve and ray Or curve equality constraint when its reverse extension is tangent: in, Represents rays The slope of Represents rays and coordinate system The intercept of the axis; Combined with the intersection point equality constraint, the incident angle of the end point of the double arc reference curve is calculated ; Double arc datum curve passes through double ray endpoints Curve equality constraints when : Combined with the intersection point equality constraint, the incident angle of the end point of the double arc reference curve is calculated ; Define the maximum incident angle at the end point of the double arc reference curve : in, Indicates the incident angle of the end point of the double arc reference curve when the intersection point is at the extreme position of the intersection point. Represents double arc datum curve and ray Or the incident angle at the end point when its reverse extension line is tangent, Indicates that the double arc datum curve passes through the ray endpoint The end point incident angle at ; Define the minimum incident angle at the end point of the double arc reference curve : in, Indicates the incident angle of the end point of the double arc reference curve when it passes through the corner of the structured step terrain; Establish the incident angle at the end point of the double arc reference curve Inequality constraints: The incident angle of the end point of the selected double arc reference curve is ; The control point spacing non-collision inequality constraint means: in, Indicates the spacing between control points on the second arc segment. Indicates the radius of the second arc, Indicates the distance from the center of the second arc segment to the corner of the structured stepped terrain.

8. The robot multi-constraint motion planning system according to claim 6, wherein: The multiple constraints of the front, rear and waist joint position trajectories of the first and second quadruped robots are established for each obstacle crossing stage, wherein the second stage is the climbing stage of the front legs of the first quadruped robot, which belongs to the movement stage; in the second stage, the position parameters of the octapod robot body are The direction remains unchanged, along Change in position per unit time of direction , is the period of the swing phase; keeping the height of the first quadruped robot body unchanged, sequentially setting the left front leg and the right front leg of the first quadruped robot as climbing legs, and using the climbing leg foot end swing trajectory to perform the climbing action until the left front leg and the right front leg of the first quadruped robot both climb onto the structured step terrain; The multiple constraints of the front, rear and waist joint position trajectories of the first and second quadruped robots are established for each obstacle crossing stage, wherein the third stage is the climbing stage of the hind legs of the first quadruped robot. The third stage belongs to the movement stage. The position parameters of the octapod robot body are The direction remains unchanged, along Change in position per unit time of direction Adopting a static gait and a non-climbing leg-foot end swing trajectory, and adjusting the leg-foot end positions of the octapod robot according to the non-climbing state step length inequality constraint, the left and right hind legs of the first quadruped robot ultimately satisfy the climbing distance inequality constraint and the climbing state step length inequality constraint; The left hind leg and the right hind leg of the first quadruped robot are set as climbing legs in turn, the height of the first quadruped robot body is kept unchanged, and the climbing leg foot end adopts the climbing leg foot end swing trajectory to perform the climbing action until the left hind leg and the right hind leg of the first quadruped robot both climb onto the structured step terrain.

9. The robot multi-constraint motion planning system according to claim 6, characterized in that: The multiple constraints for the position trajectories of the front, rear, and waist joints of the first and second quadruped robots are established for each obstacle crossing stage, wherein the constraints for stage four include: The position trajectory of the fuselage in this stage can be obtained by the multiple constraints of the fuselage position parameter trajectory; Phase 4: The first quadruped robot resumes its horizontal motion, and the second quadruped robot is raised. Phase 4 is the posture adjustment phase. The second quadruped robot performs the same actions as the first quadruped robot. In this phase, the first quadruped robot's body pitch angle is adjusted from Adjust to 0, the pitch angle of the second quadruped robot body is adjusted from 0 to ; The second quadruped robot rotates clockwise around point F, so the motion trajectory of point C is an arc with point F as the center and CF as the radius, the motion trajectory of point D is an arc with point F as the center and DF as the radius, and the motion trajectory of point E is an arc with point F as the center and EF as the radius.

10. The robot multi-constraint motion planning system according to claim 6, characterized in that: The multi-constraint conditions for the position trajectory of the front, rear and waist joints of the first and second quadruped robots are established for each obstacle crossing stage, wherein the fifth stage is the climbing stage of the front legs of the second quadruped robot; the fifth stage belongs to the movement stage; the position parameters of the eight-legged robot body are The direction remains unchanged, along Change in position per unit time of direction , is the period of the swing phase; a static gait and a non-climbing leg foot end swing trajectory are adopted, and the position of each leg foot end of the octapod robot is adjusted according to the non-climbing state step length inequality constraint, so that the left front leg and the right front leg of the second quadruped robot finally satisfy the climbing distance inequality constraint and the climbing state step length inequality constraint; the left front leg and the right front leg of the second quadruped robot are set as climbing legs in turn, and the height of the second quadruped robot body is kept unchanged. The climbing leg foot end adopts the climbing leg foot end swing trajectory to perform the climbing action until the left front leg and the right front leg of the second quadruped robot both climb onto the structured step terrain; The multi-constraint conditions for the front, rear and waist joint position trajectories of the front and rear quadruped robot bodies are established for each obstacle crossing stage, wherein the sixth stage is the second quadruped robot hind leg climbing stage, the sixth stage belongs to the movement stage, and the eight-legged robot body position parameters are The direction remains unchanged, along Change in position per unit time of direction ; A static gait and a non-climbing leg foot end swing trajectory are adopted, and the position of each leg foot end of the eight-legged robot is adjusted according to the non-climbing state step length inequality constraint, so that the left hind leg and the right hind leg of the second quadruped robot finally meet the climbing distance inequality constraint and the climbing state step length inequality constraint; the left hind leg and the right hind leg of the second quadruped robot are set as climbing legs in turn, and the height of the second quadruped robot body is kept unchanged. The climbing leg foot end adopts the climbing leg foot end swing trajectory to perform the climbing action until the left hind leg and the right hind leg of the second quadruped robot both climb onto the structured step terrain; The multiple constraints for the position trajectories of the front, rear, and waist joints of the front and rear quadruped robots are established for each obstacle crossing stage, wherein the constraints for stage seven include: The position trajectory of the fuselage in this stage can be obtained by the multiple constraints of the fuselage position parameter trajectory; in, Indicates the height of structured step terrain; Phase 7: The second quadruped robot resumes its translation phase. Phase 7 is the posture adjustment phase. In Phase 7, the first quadruped robot body pitch angle is kept at 0, and the second quadruped robot body pitch angle is changed from Adjust to 0; first keep the position of point C unchanged, the second quadruped robot rotates counterclockwise around point C until the pitch angle of the second quadruped robot body is 0, and the first quadruped robot maintains the height of the translation stage unchanged; the trajectory of point D is an arc with C as the center and CD as the radius, and the trajectory of point E is an arc with C as the center and CE as the radius.

11. A modular octapod robot, characterized in that: The robot includes a processor and a memory, wherein the memory stores at least one instruction, at least one program, a code set or an instruction set, and the at least one instruction, the at least one program, the code set or the instruction set is loaded and executed by the processor to implement the planning method described in any one of claims 1 to 5 above.

12. A computer-readable storage medium, characterized in that: The storage medium stores at least one instruction, at least one program, code set or instruction set, and the at least one instruction, at least one program, code set or instruction set is loaded and executed by the processor to implement the planning method described in any one of claims 1 to 5 above.

Images