Optimization method for structural parameters of leg mechanism of unmanned metamorphic vehicle

Abstract

The invention discloses an optimization method for structural parameters of a leg mechanism of an unmanned metamorphic vehicle, and the method comprises the steps: 1, building a kinematics model and a dynamics model of the unmanned metamorphic vehicle based on a homogeneous coordinate transformation method and a Lagrange theory, 2, taking ZMP as a stability index, 3, setting a thigh length and a shank length as optimization variables, and setting constraint conditions, the method comprises the steps of establishing a multi-objective function containing motion energy consumption and motion stability, and converting the multi-objective function into a single-objective function through a linear weighting method, and 4, optimizing the leg length by using an ant colony algorithm, constructing a crawling path and nodes, setting an initial ant population and related parameters, calculating the ant crawling probability, and updating the pheromone concentration until a termination condition is met. According to the method, the stability of the unmanned metamorphic vehicle during reconstruction and walking can be remarkably improved, for example, the xZMP curve is far away from a support domain boundary during reconstruction, and the deviation from an ideal curve during walking is reduced; the movement energy consumption is also greatly reduced, and the joint energy consumption in the reconstruction and walking processes is obviously reduced.

Description

Technical Field

[0001] The present invention belongs to the technical field of unmanned metamorphic vehicles, and particularly relates to the optimization of structural parameters of a leg mechanism of an unmanned metamorphic vehicle in its design. Background Art

[0002] The unmanned metamorphic vehicle is a new type of reconstructed robot developed based on the structure of an automobile. It has multiple movement modes. It can achieve high-speed and stable driving in the form of an automobile on structured roads. When encountering unstructured roads (steps, stairs, etc.), it can also achieve human-like striding with the support of its legs and feet. The vehicle can be widely used in engineering surveys, military reconnaissance, anti-terrorism and riot prevention, and interstellar exploration.

[0003] However, the process of reconfiguring from a car-like state to a humanoid state presents numerous challenges for unmanned transformable vehicles. For one thing, the center of mass position changes significantly, and the support domain shrinks from the area enclosed by four wheels to the area enclosed by two feet. This significant reduction in support area makes it highly susceptible to tipping and instability during the reconfiguration process. Furthermore, research into reducing robot motion energy consumption and improving stability has largely focused on reducing leg mass, a relatively simple approach that fails to effectively improve robot stability. Furthermore, the length and ratio of the thigh and shank of the leg structure affect the stability and energy consumption of the robot's reconfiguration motion. Therefore, optimizing the optimal leg length to improve reconfiguration stability and reduce energy consumption has become a pressing challenge. Summary of the Invention

[0004] The purpose of the present invention is to provide a method for optimizing the structural parameters of the leg mechanism of an unmanned metamorphic vehicle, so as to optimize the leg length structural parameters of the leg mechanism of the unmanned metamorphic vehicle, thereby comprehensively considering the leg length structural parameters when the leg mechanism moves, reducing energy consumption and improving the stability of movement.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0006] The present invention provides a method for optimizing the structural parameters of a leg mechanism of an unmanned metamorphic vehicle. When the unmanned metamorphic vehicle walks in a humanoid state, one leg in the leg mechanism serves as a supporting leg, the other leg serves as a swinging leg, and the ankle joint of the swinging leg serves as the terminal mechanism. When the unmanned metamorphic vehicle undergoes a reconstructed motion of rotating between a humanoid state and a vehicle state, both legs in the leg mechanism serve as supporting legs, and the hip joint of the supporting leg serves as the terminal mechanism. The method is characterized in that the structural parameter optimization comprises the following steps:

[0007] Step 1. Establish the kinematic model and dynamic model of the unmanned metamorphic vehicle;

[0008] Step 2: Take the length of the thigh as the , calf length To optimize variables, set the constraints of the optimized variables and the objective function J of the optimized variables;

[0009] Step 3: Optimize the optimization variables based on the ant colony algorithm to obtain the optimal variables, including the optimal leg length of the thigh Optimal calf length .

[0010] The structural parameter optimization method of the leg mechanism of the unmanned metamorphic vehicle described in the present invention is also characterized in that the step 1 comprises:

[0011] Step 1.1. Establish the kinematic model of the unmanned metamorphic vehicle based on the homogeneous coordinate transformation method;

[0012] Construct any i-1 joint coordinate system among the ankle, knee, and hip joints in the leg mechanism ; The angle of clockwise rotation around the coordinate axis is positive, and the opposite is negative; ; Indicates the number of components, and n=6;

[0013] Assume the i-1th joint coordinate system Around The pitch rotation angle of the axis is , then along Axis translation , and then along Axis translation Then, we get the i-th joint coordinate system ;

[0014] Using formula (1) to get the i-th joint coordinate system Relative to the i-1th joint coordinate system The coordinate transformation matrix , thus obtaining the homogeneous transformation matrix of each joint coordinate system relative to the basic coordinate system in turn according to formula (1), including: the homogeneous transformation matrix of the right ankle joint coordinate system relative to the basic coordinate system , the homogeneous transformation matrix of the right knee joint coordinate system relative to the right ankle coordinate system , the homogeneous transformation matrix of the right hip joint coordinate system relative to the right knee joint coordinate system , the homogeneous transformation matrix of the left hip joint coordinate system relative to the right hip joint coordinate system , the homogeneous transformation matrix of the left knee joint coordinate system relative to the left hip joint coordinate system , the homogeneous transformation matrix of the left ankle joint coordinate system relative to the left knee joint coordinate system , where the origin of the base coordinate system is at the center of the supporting foot;

[0015] (1)

[0016] Formula (2) is used to obtain the homogeneous transformation matrix of each joint coordinate system relative to the basic coordinate system during walking, thereby obtaining the position coordinates of the center of mass of each component relative to the basic coordinate system during walking. Among them, the coordinate transformation matrix of the ankle joint of the swinging leg relative to the basic coordinate system during walking is :

[0017] (2)

[0018] Formula (3) is used to obtain the homogeneous transformation matrix of each joint coordinate system relative to the basic coordinate system during the reconstruction process, thereby obtaining the position coordinates of the center of mass of each component relative to the basic coordinate system during the reconstruction process. Among them, the coordinate transformation matrix of the hip joint coordinate system of any supporting leg relative to the basic coordinate system during the reconstruction process is ;

[0019] (3)

[0020] Step 1.2. Establish a dynamic model of the unmanned metamorphic vehicle based on Lagrangian dynamics;

[0021] According to the analytical mechanics theory, the Lagrangian dynamics model of the leg mechanism of the unmanned metamorphic vehicle during walking and reconstruction is established using formula (4):

[0022] (4)

[0023] In formula (4), is with The relevant generalized force, for The derivative of represents the integral, Indicates time, The kinetic energy of the unmanned transforming car, The potential energy of the unmanned transforming car, is the Lagrangian operator of the unmanned vehicle; and:

[0024] (5)

[0025] (6)

[0026] In formula (5) and formula (6), for The derivative of express No. elements, represents transpose; represents the position coordinates of the center of mass of the i-th component in the basic coordinate system, represents the mass of the i-th component;

[0027] Step 1.3. Use the zero moment point theory ZMP as the stability criterion of the unmanned metamorphic vehicle and use formula (7) to establish the stability index of the unmanned metamorphic vehicle in the X-axis direction during the walking process and reconstruction motion. And the stability index of the Y axis direction ;

[0028] (7)

[0029] In formula (7), represents the second-order derivative of the Z-axis coordinate of the i-th component in the basic coordinate system, represents the second-order derivative of the X-axis coordinate of the i-th component in the basic coordinate system, Indicates the Z-axis coordinate of the i-th component in the basic coordinate system, represents the Y-axis coordinate of the i-th component in the basic coordinate system, Represents the acceleration due to gravity.

[0030] Furthermore, the step 2 includes:

[0031] Step 2.1: Set the constraints of the optimization variables;

[0032] Formula (8) is used to establish the constraints of the unmanned metamorphic vehicle under reconstruction motion:

[0033] (8)

[0034] In formula (8), and Represent the angles of the ankle and knee joints at the start of reconstruction, Indicates the length of the sole of the foot. Indicates the distance between the leg mechanism and the front extreme position of the vehicle body during the deployment process under the reconstruction movement, Indicates the height from the ground to the center of rotation of the ankle joint around the Y axis in its own joint coordinate system. Indicates the height between the rotation center of the Y-axis actuator of the hip joint and the mounting plate of the leg mechanism. Indicates the height between the mounting plate of the leg mechanism and the horizontal ground in the vehicle state;

[0035] Formula (9) is used to establish the constraints of the unmanned vehicle during its movement:

[0036] (9)

[0037] In formula (9), represents the center of mass height planned by the inverted pendulum model during walking;

[0038] Formula (10) is used to establish the leg length ratio constraint of the unmanned metamorphic vehicle:

[0039] (10)

[0040] In formula (10), Indicates the minimum ratio of thigh length to calf length of the leg mechanism; Indicates the maximum ratio of the thigh length to the calf length of the leg mechanism;

[0041] Step 2.2: Establish the objective function of the optimization variables, including: motion energy consumption function and motion stability function;

[0042] Formula (11) is used to establish the power consumed by the i-th joint of the unmanned metamorphic vehicle during walking or reconstruction motion: :

[0043] (11)

[0044] Formula (12) is used to establish the total energy consumption objective function of all joints of the unmanned metamorphic vehicle during walking. And the total energy consumption objective function of all joints in the reconstruction movement :

[0045] (12)

[0046] In formula (12), represents the energy consumed by the i-th joint during walking or reconstruction movement; Indicates the total duration of the walking process, represents the total duration of the motion during the reconstruction process;

[0047] Let the expected zero moment point ZMP in the X direction during the motion of the unmanned vehicle be x,d =0, the desired zero moment point in the Y direction is ZMP y,d ; Thus, the stability objective function of the unmanned metamorphic vehicle during the walking process is determined using formula (13) and the stability objective function of all joints in the reconstructed motion ;

[0048] (13)

[0049] In formula (13), Indicates the true zero moment point in the X direction; Indicates the true zero moment point in the Y direction;

[0050] Using formula (14), we can get the single objective function in the walking process and the reconstruction motion: :

[0051] (14)

[0052] In formula (14), 、 、 、 Represents 4 weights.

[0053] Furthermore, the step 3 includes:

[0054] Step 3.0: Define the current number of iterations as g and initialize g=1; the maximum number of iterations is ;

[0055] Let the total number of ants in the ant population be m, and each ant individual represents a set of optimization variables;

[0056] Step 3.1: Initialize the ant number k=1;

[0057] Define that any k-th ant in the g-th generation ant population contains M genes. The first M / 2 genes represent the leg length of the thigh, and the last M / 2 genes represent the leg length of the calf. Denotes the s-th gene of any k-th ant in the g-th generation ant population, let The value of ,Depend on The sth node coordinates of any kth ant in the gth generation ant population;

[0058] Randomly initialize the pheromone concentration on the path connecting the sth node coordinate and the s+1th node coordinate shared by all ants in the gth generation ant population ;

[0059] Step 3.2: Initialize s=1; in the feasible interval [ , ] randomly initialize the value of each gene of the k-th ant in the g-th generation ant population, and form the s-th node coordinate of the k-th ant in the g-th generation ant population; randomly select an ant in the g-th generation ant population as the optimal ant, and let the value of its s-th gene be ;

[0060] Step 3.3: Use formula (15) to calculate the visibility information of any k-th ant in the g-th generation ant population on the connection path between the s-th node coordinate and the s+1-th node coordinate :

[0061] (15)

[0062] In formula (15), represents the number of any k-th ant in the g-th generation ant population. The value of the bit gene, Indicates the best ant in the g-1 generation. The value of the bit gene, ( ) represents the set of all next feasible node coordinates of the sth node coordinate of the kth ant in the gth iteration ( ) in the feasible node coordinates;

[0063] Step 3.4: Use formula (16) to calculate the distance that any k-th ant in the g-th generation ant population crawls from the s-th node coordinate to ( ) in the state transition probability of the s+1th node coordinate , thus according to , take the gene value corresponding to the selected s+1th node coordinate As the kth ant in the gth generation of ants, alleles The updated value;

[0064] (16)

[0065] In formula (16), represents the pheromone concentration on the connection path between the sth node coordinate and the s+1th node coordinate shared by all ants in the gth iteration, is the visibility information on the connection path between the coordinates of the sth node and the coordinates of the s+1th node in the kth ant at the gth iteration, and They are pheromone factor and visibility factor respectively;

[0066] Step 3.5: After assigning s+1 to s, if s>S, the updated gene value of each gene of the k-th ant in the g-th generation ant population is obtained, and step 3.5 is executed; otherwise, return to step 3.3 and execute sequentially;

[0067] Step 3.6: After assigning k+1 to k, if k>m, the updated gene values ​​of each gene of all ants in the g-th generation ant population are obtained. The ant with the smallest objective function value in the g-th generation ant population is calculated as the optimal ant of the g-th generation, and step 3.7 is executed. Otherwise, return to step 3.2 and execute sequentially.

[0068] Step 3.7: Determine whether the objective function value of the best ant of generation g is less than that of the best ant of generation g-1. If so, the best ant of generation g remains unchanged. Otherwise, the best ant of generation g-1 is assigned to the best ant of generation g.

[0069] Step 3.8: Determine whether g = G. If so, stop the iteration and output the optimal ant of the Gth generation ant population as the optimal thigh length and optimal calf length. Otherwise, proceed to step 3.9.

[0070] Step 3.9: Use formula (17) to calculate the pheromone increment of any k-th ant in the g-th generation ant population crawling from the s-th node coordinate to the s+1-th node coordinate :

[0071] (17)

[0072] In formula (17), is the pheromone intensity, is the objective function of any k-th ant in the g-th generation ant population. When g=1, let ;

[0073] Step 3.10: Use formula (18) to calculate the sum of pheromone increments on the connection path between the sth node coordinate and the s+1th node coordinate shared by all ants in the gth generation ant population. , thus calculating the The pheromone concentration shared by all ants in the generation ant population crawling from the sth node coordinate point to the s+1th node coordinate point :

[0074] (18)

[0075] In formula (18), is a constant that reflects the evaporation rate of pheromones, and <1;

[0076] Step 3.11: After assigning g+1 to g, step 3.2 is executed sequentially.

[0077] The electronic device of the present invention includes a memory and a processor, and is characterized in that the memory is used to store a program that supports the processor to execute the optimization method, and the processor is configured to execute the program stored in the memory.

[0078] The present invention provides a computer-readable storage medium, wherein a computer program is stored on the computer-readable storage medium, and the computer program executes the steps of the optimization method when executed by a processor.

[0079] Compared with the prior art, the beneficial effects of the present invention are embodied in:

[0080] 1. This invention establishes the kinematics, dynamics model and stability index of the unmanned metamorphic vehicle, and uses the ant colony algorithm to optimize the thigh and calf lengths of the leg mechanism, thus significantly improving the stability of the unmanned metamorphic vehicle during reconstruction and walking. The curve is farther away from the support area boundary during the whole process, which reduces the risk of tipping and instability. The deviation between the curve and the ideal curve becomes smaller, and the walking stability is effectively improved.

[0081] 2. This invention optimizes the structural parameters of the unmanned metamorphic vehicle's leg mechanism, comprehensively considering the energy consumption of the leg mechanism during movement. After optimization, joint energy consumption during reconstruction was reduced by 71.703 J, a 19.11% reduction compared to the pre-optimization process; energy consumption during walking was reduced by 18.175 J, a 15.00% reduction compared to the pre-optimization process, effectively reducing the unmanned metamorphic vehicle's energy consumption during movement. BRIEF DESCRIPTION OF THE DRAWINGS

[0082] Figure 1 This is a kinematic diagram of the connecting rods of the leg mechanism of the unmanned metamorphic vehicle involved in the present invention;

[0083] Figure 2 Schematic diagram of variable constraints for optimizing the structural parameters of the leg mechanism of the unmanned metamorphic vehicle of the present invention;

[0084] Figure 3 This is a flow chart of the algorithm for optimizing the structural parameters of the leg mechanism of the unmanned metamorphic vehicle involved in the present invention;

[0085] Figure 4 A comparison diagram of the structural parameters of the leg mechanism of the unmanned metamorphic vehicle of the present invention before and after the optimization of the reconstruction process;

[0086] Figure 5 This is a comparison diagram of the leg mechanism structural parameters of the unmanned metamorphic vehicle involved in the present invention before and after optimization during the walking process. DETAILED DESCRIPTION

[0087] In this embodiment, when the unmanned metamorphic vehicle is walking in a humanoid state, one leg in the leg mechanism serves as a supporting leg, the other leg serves as a swinging leg, and the ankle joint of the swinging leg serves as the end mechanism; when the unmanned metamorphic vehicle is in a reconstructed movement of rotating between the humanoid state and the automobile state, both legs in the leg mechanism serve as supporting legs, and the hip joint of the supporting leg serves as the end mechanism.

[0088] A method for optimizing the structural parameters of the leg mechanism of an unmanned metamorphic vehicle comprises the following steps:

[0089] Step 1. Establish the kinematic model and dynamic model of the unmanned metamorphic vehicle;

[0090] Step 1.1. Establish the kinematic model of the unmanned metamorphic vehicle based on the homogeneous coordinate transformation method;

[0091] Construct any i-1 joint coordinate system among the ankle, knee, and hip joints in the leg mechanism ; The angle of clockwise rotation around the coordinate axis is positive, and the opposite is negative; ; Indicates the number of components, and n=6;

[0092] Assume the i-1th joint coordinate system Around The pitch rotation angle of the axis is , then along Axis translation , and then along Axis translation Then, we get the i-th joint coordinate system ;

[0093] Using formula (1) to get the i-th joint coordinate system Relative to the i-1th joint coordinate system The coordinate transformation matrix , thus obtaining the homogeneous transformation matrix of each joint coordinate system relative to the basic coordinate system in turn according to formula (1), including: the homogeneous transformation matrix of the right ankle joint coordinate system relative to the basic coordinate system , the homogeneous transformation matrix of the right knee joint coordinate system relative to the right ankle coordinate system , the homogeneous transformation matrix of the right hip joint coordinate system relative to the right knee joint coordinate system , the homogeneous transformation matrix of the left hip joint coordinate system relative to the right hip joint coordinate system , the homogeneous transformation matrix of the left knee joint coordinate system relative to the left hip joint coordinate system , the homogeneous transformation matrix of the left ankle joint coordinate system relative to the left knee joint coordinate system , where the origin of the base coordinate system is at the center of the supporting sole, as Figure 1 As shown, Figure 1 Part (a) shows the 3D model of the leg mechanism. Figure 1 Part (b) is the simplified link mechanism of the leg mechanism components, and the origin of each coordinate system is established at the center of each joint.

[0094] Joint coordinate system Relative to The homogeneous transformation matrix for:

[0095] (1)

[0096] Formula (2) is used to obtain the homogeneous transformation matrix of each joint coordinate system relative to the basic coordinate system during walking, thereby obtaining the position coordinates of the center of mass of each component relative to the basic coordinate system during walking. Among them, the coordinate transformation matrix of the ankle joint of the swinging leg relative to the basic coordinate system during walking is :

[0097] (2)

[0098] Formula (3) is used to obtain the homogeneous transformation matrix of each joint coordinate system relative to the basic coordinate system during the reconstruction process, thereby obtaining the position coordinates of the center of mass of each component relative to the basic coordinate system during the reconstruction process. Among them, the coordinate transformation matrix of the hip joint coordinate system of any supporting leg relative to the basic coordinate system during the reconstruction process is ;

[0099] (3)

[0100] Step 1.2. Establish a dynamic model of the unmanned metamorphic vehicle based on Lagrangian dynamics;

[0101] According to the analytical mechanics theory, the Lagrangian dynamic model of the leg mechanism of the unmanned metamorphic vehicle during walking and reconstruction is established using formula (4):

[0102] (4)

[0103] In formula (4), is with The relevant generalized force ( are the rotation angles of the ankle, knee, and hip joints of the right leg and the hip, knee, and ankle joints of the left leg), for The derivative of represents the integral, Indicates time, The kinetic energy of the unmanned transforming car, The potential energy of the unmanned transforming car, is the Lagrangian operator of the unmanned vehicle; and:

[0104] (5)

[0105] In formula (5), is a 6×6 mass matrix, let is the mass matrix No. elements, is a 6×1 vector consisting of 6 joint angular velocities, express No. , elements.

[0106] (6)

[0107] In formula (5) and formula (6), for The derivative of express No. elements, represents transpose; , =9.8m / s 2 , represents the position coordinates of the center of mass of the i-th component in the basic coordinate system, represents the mass of the i-th component.

[0108] Step 1.3. Use the zero moment point theory ZMP as the stability criterion of the unmanned metamorphic vehicle and use formula (7) to establish the stability index of the unmanned metamorphic vehicle in the X-axis direction during the walking process and reconstruction motion. And the stability index of the Y axis direction ;

[0109] (7)

[0110] In formula (7), is the acceleration due to gravity, For components quality, represents the second-order derivative of the Z-axis coordinate of the i-th component in the basic coordinate system, represents the second-order derivative of the X-axis coordinate of the i-th component in the basic coordinate system, Indicates the Z-axis coordinate of the i-th component in the basic coordinate system, represents the Y-axis coordinate of the i-th component in the basic coordinate system, Represents the acceleration due to gravity.

[0111] Step 2: Take the length of the thigh as the , calf length To optimize the variables, set the constraints of the optimized variables and establish the objective function of the optimized variables based on the kinematic and dynamic models of the leg mechanism.

[0112] Step 2.1: Set the constraints of the optimization variables;

[0113] The optimization variables of the leg mechanism are: thigh length , calf length , based on the initial design value For the range of variation. Figure 2 As shown, according to the constraints of the reconstruction working condition, when the car state is transformed into the support state, it is necessary to ensure that the leg mechanism can be fully supported on the ground, that is, the vertical height of the leg mechanism in the support state is not less than the height from the leg mounting plate to the horizontal ground in the car state. In addition, it is necessary to ensure that the leg mechanism is in the front and rear extreme positions of the vehicle body during the deployment process. 、 No collision or contact occurs. For the reconstruction workcase, the thigh and calf sizes are constrained.

[0114] Formula (8) is used to establish the constraints of the unmanned metamorphic vehicle under reconstruction motion:

[0115] (8)

[0116] In formula (8), and Represent the angles of the ankle and knee joints at the start of reconstruction, Indicates the length of the sole of the foot. Indicates the distance between the leg mechanism and the front extreme position of the vehicle body during the deployment process under the reconstruction movement, Indicates the height from the ground to the center of rotation of the ankle joint around the Y axis in its own joint coordinate system. Indicates the height between the rotation center of the Y-axis actuator of the hip joint and the mounting plate of the leg mechanism. Indicates the height between the mounting plate of the leg mechanism and the horizontal ground in the vehicle state.

[0117] During walking, when the unmanned vehicle is fully standing, it is necessary to ensure that the height of the leg mechanism from the ground is less than the center of mass height planned by the inverted pendulum model. 0.9m, with restraints on the thigh and calf;

[0118] Formula (9) is used to establish the constraints of the unmanned vehicle during its movement:

[0119] (9)

[0120] In formula (9), represents the center of mass height planned by the inverted pendulum model during walking;

[0121] Some well-known bipedal robots, such as TORO and WALK-MAN, have thighs and calves of roughly equal length, with a 1:1 ratio. However, the thighs of the leg mechanism in this paper need to fit into the box-shaped slots of the calves during folding and contraction, so the calves are designed to be slightly longer than the thighs. At the same time, the difference in thigh and calf length should not be too large. Otherwise, an excessively long calf will increase its mass, which in turn will increase the moment of inertia of the calf around the knee joint and increase joint energy consumption. The set thigh-to-calf length ratio constraint;

[0122] Formula (10) is used to establish the leg length ratio constraint of the unmanned metamorphic vehicle:

[0123] (10)

[0124] In formula (10), Indicates the minimum ratio of the thigh length to the calf length of the leg mechanism, ; Indicates the maximum ratio of the thigh length to the calf length of the leg mechanism, .

[0125] Step 2.2: Establish the objective function of the optimization variables;

[0126] This paper is a multi-objective optimization problem, and the optimization objective functions set include motion energy consumption function and motion stability function. Among them, energy consumption mainly refers to the mechanical energy consumption of the joint motor during the movement process;

[0127] Formula (11) is used to establish the power consumed by the i-th joint of the unmanned metamorphic vehicle during walking or reconstruction motion: :

[0128] (11)

[0129] Formula (12) is used to establish the total energy consumption objective function of all joints of the unmanned metamorphic vehicle during walking. And the total energy consumption objective function of all joints in the reconstruction movement :

[0130] (12)

[0131] In formula (12), represents the energy consumed by the i-th joint during walking or reconstruction movement; represents the total motion duration of the reconstruction process, Indicates the total duration of the walking process.

[0132] Let the expected zero moment point ZMP in the X direction during the motion of the unmanned vehicle be x,d=0, the desired zero moment point in the Y direction is ZMP y,d ; Thus, the stability objective function of the unmanned metamorphic vehicle during the walking process is determined using formula (13) and the stability objective function of all joints in the reconstructed motion ;

[0133] (13)

[0134] In formula (13), Indicates the true zero moment point in the X direction; Indicates the true zero moment point in the Y direction;

[0135] For the reconstruction condition, the total reconstruction time is set to The energy consumption during the reconstruction process mainly comes from the mechanical legs, the center of mass adjustment mechanism, and the horizontal lifting mechanism. Compared with the mechanical legs, the other mechanisms consume very little energy during the reconstruction process, so only the energy consumption optimization of the mechanical legs is considered. changes, Will not change, so the total energy consumption of all joints during the reconstruction process and stability goals They are:

[0136] (14)

[0137] For the walking condition, the total movement time is set to The energy consumption during walking mainly comes from the mechanical legs and the center of mass adjustment mechanism. Compared with the mechanical legs, the center of mass adjustment mechanism consumes very little energy during walking and can be ignored. The forward movement of the legs during walking causes the system to Change. And It is mainly regulated by adjusting the movement of the center of mass slider and has nothing to do with the leg mechanism. and stability goals for:

[0138] (15)

[0139] Using formula (16), we can get the single objective function in the walking process and the reconstruction motion: :

[0140] (16)

[0141] In formula (16), 、 、 、 Represents 4 weights, and the multi-objective function of stability and energy consumption under walking and reconstruction conditions is transformed into a single objective function through the linear weighting method In both conditions, stability takes precedence over energy consumption. Stability has equal priority in both walking and reconstruction conditions; since reconstruction motion consumes more energy than walking, energy consumption takes precedence in reconstruction motion over walking motion. 、 、 、 .

[0142] Step 3: Optimize the optimization variables based on the ant colony algorithm to obtain the optimal variables, including the optimal leg length of the thigh Optimal calf length ;

[0143] The ant colony optimization algorithm is a bionic intelligent optimization algorithm that simulates the collective foraging behavior of ants in nature. It has the advantages of strong robustness, parallel distributed computing and fast convergence speed. The ant colony optimization algorithm is used to optimize the parameters in the objective function. The algorithm process is as follows: Figure 3 shown.

[0144] Step 3: Optimize the optimization variables based on the ant colony algorithm to obtain the optimal variables, including the optimal leg length of the thigh Optimal calf length ;

[0145] Step 3.0: Define the current number of iterations as g and initialize g=1; the maximum number of iterations is ;

[0146] Let the total number of ants in the ant population be m, and each ant individual represents a set of optimization variables;

[0147] Step 3.1: Initialize the ant number k=1;

[0148] Define that any k-th ant in the g-th generation ant population contains M genes. The first M / 2 genes represent the leg length of the thigh, and the last M / 2 genes represent the leg length of the calf. Denotes the s-th gene of any k-th ant in the g-th generation ant population, let The value of ,Depend on The sth node coordinates of any kth ant in the gth generation ant population;

[0149] Randomly initialize the pheromone concentration on the path connecting the sth node coordinate and the s+1th node coordinate shared by all ants in the gth generation ant population .

[0150] Step 3.2: Initialize s=1; in the feasible interval [ , ] is randomly initialized to obtain the value of each gene of the k-th ant in the g-th generation ant population, and constitute the s-th node coordinate of the k-th ant in the g-th generation ant population;

[0151] Randomly select an ant from the g-th generation ant population as the best ant, and let the value of its s-th gene be ;

[0152] Step 3.3: Use formula (17) to calculate the visibility information of any k-th ant in the g-th generation ant population on the connection path between the s-th node coordinate and the s+1-th node coordinate :

[0153] (17)

[0154] In formula (17), represents the number of any k-th ant in the g-th generation ant population. The value of the bit gene, Indicates the best ant in the g-1 generation. The value of the bit gene, ( ) represents the set of all next feasible node coordinates of the sth node coordinate of the kth ant in the gth iteration ( ) in the feasible node coordinates; when =1, let = ;

[0155] Step 3.4: Use formula (18) to calculate the distance that any k-th ant in the g-th generation ant population crawls from the s-th node coordinate to ( ) in the state transition probability of the s+1th node coordinate , thus according to , take the gene value corresponding to the selected s+1th node coordinate As the kth ant in the gth generation of ants, alleles The updated value;

[0156] (18)

[0157] In formula (18), represents the pheromone concentration on the connection path between the sth node coordinate and the s+1th node coordinate shared by all ants in the gth iteration, is the visibility information on the connection path between the coordinates of the sth node and the coordinates of the s+1th node in the kth ant at the gth iteration, and They are pheromone factor and visibility factor respectively.

[0158] Step 3.5: After assigning s+1 to s, if s>S, the updated gene value of each gene of the k-th ant in the g-th generation ant population is obtained, and step 3.5 is executed; otherwise, return to step 3.3 and execute sequentially;

[0159] Step 3.6: After assigning k+1 to k, if k>m, the updated gene values ​​of each gene of all ants in the g-th generation ant population are obtained. The ant with the smallest objective function value in the g-th generation ant population is calculated as the optimal ant of the g-th generation, and step 3.7 is executed. Otherwise, return to step 3.2 and execute sequentially.

[0160] Step 3.7: Determine whether the objective function value of the best ant of generation g is less than that of the best ant of generation g-1. If so, the best ant of generation g remains unchanged. Otherwise, the best ant of generation g-1 is assigned to the best ant of generation g.

[0161] Step 3.8: Determine whether g = G. If so, stop the iteration and output the optimal ant of the G-th generation ant population as the optimal thigh length and optimal calf length. Otherwise, proceed to step 3.9.

[0162] Step 3.9: Use formula (19) to calculate the pheromone increment of any k-th ant in the g-th generation ant population crawling from the s-th node coordinate to the s+1-th node coordinate :

[0163] (19)

[0164] In formula (19), is the pheromone intensity, is the objective function of any k-th ant in the g-th generation ant population. When g=1, let .

[0165] Step 3.10: Use formula (20) to calculate the sum of pheromone increments on the connection path between the sth node coordinate and the s+1th node coordinate shared by all ants in the gth generation ant population. , thus calculating the The pheromone concentration shared by all ants in the generation ant population crawling from the sth node coordinate point to the s+1th node coordinate point :

[0166] (20)

[0167] In formula (20), is a constant that reflects the evaporation rate of pheromones, and <1.

[0168] Step 3.11: After assigning g+1 to g, step 3.2 is executed sequentially.

[0169] In this embodiment, an electronic device includes a memory and a processor, wherein the memory is used to store a program that supports the processor to execute the above method, and the processor is configured to execute the program stored in the memory.

[0170] In this embodiment, a computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the above method are executed.

[0171] Example: Set the ant population number m=10, and They are 3 and 5 respectively. is 0.7, the pheromone strength =1, the pheromone concentration at the initial moment , in the first generation, an ant is randomly selected as the optimal ant, and the maximum number of iterations is In the reconstruction condition, the reconstruction time was set to 8 s, and in the walking condition, the walking time was set to 1.2 s (one left single-foot support period, one right single-foot support period, and two double-foot support periods), with a step length of 0.4 m.

[0172] According to the motion planning of each joint, the comparison results of the objective function before and after optimization are shown in Table 1. The energy consumption of the joints in the reconstruction process after optimization is Reduced by 71.703J, a decrease of 19.11% compared to before optimization; stability The energy consumption of the walking process after optimization was reduced by 18.175J, which was 15.00% lower than that before optimization. It decreased by 0.005, which is a 20.83% decrease compared to before optimization.

[0173] Table 1 Comparison before and after optimization

[0174]

[0175] A detailed comparison of the reconstruction process before and after optimization is shown below. Figure 4 As shown. Figure 4 In part (a) of the optimization The curve remains in the support domain throughout the reconstruction process. Although the motion stability during the reconstruction process can be guaranteed, at the end of the reconstruction at 8 seconds, the The curve is very close to the boundary of the support region, and the system has the potential danger of overturning and becoming unstable. The curve can be kept far away from the support domain boundary during the entire reconstruction process, which greatly improves the reconstruction stability. Figure 4 In part (b), because the left and right legs move synchronously during the reconstruction process, the energy consumption of the joints at the same joint locations in the left and right legs is the same. Joints 1-6 represent the ankle, knee, and hip joints of the left leg, and the hip, knee, and ankle joints of the right leg, respectively. After optimization, energy consumption of all joints decreases. The knee joint consumes the most energy during the reconstruction process, while the hip joint consumes the least energy.

[0176] In addition to being used for reconstructed motion, the leg mechanism also needs to be used for walking. During walking, the motion patterns of each joint are different from those of the reconstructed state, so the motion stability and energy consumption during walking are different from those of the reconstructed state. Figure 5 As shown. Figure 5 In part (a), before optimization, Curves and ideals The deviation of the curve is large. After optimization, this deviation has been greatly improved. Curves and ideals The deviation of the curve becomes smaller and the walking stability is effectively improved.

[0177] like Figure 5 As shown in part (b) of the figure, energy consumption at the same joints in the left and right legs is consistent. This is because both legs experience the same amount of time in the swing and stance phases. Energy consumption at the knee joint is higher than at the hip and ankle joints, with the ankle joint experiencing the lowest energy consumption. Energy consumption at all joints has decreased significantly.

Claims

1. A method for optimizing the structural parameters of the leg mechanism of an unmanned metamorphic vehicle, wherein when the unmanned metamorphic vehicle walks in a humanoid state, one leg in the leg mechanism serves as a supporting leg, the other leg serves as a swinging leg, and the ankle joint of the swinging leg serves as the terminal mechanism; when the unmanned metamorphic vehicle undergoes a reconstructed motion between a humanoid state and a vehicle state, both legs in the leg mechanism serve as supporting legs, and the hip joint of the supporting leg serves as the terminal mechanism; characterized in that: The structural parameter optimization comprises the following steps: Step 1. Establish the kinematic model and dynamic model of the unmanned metamorphic vehicle; Step 2: Take the length of the thigh as the , calf length To optimize variables, set the constraints of the optimized variables and the objective function J of the optimized variables; Step 3: Optimize the optimization variables based on the ant colony algorithm to obtain the optimal variables, including the optimal leg length of the thigh Optimal calf length .

2. The structural parameter optimization method of the leg mechanism of an unmanned metamorphic vehicle according to claim 1 is characterized in that: The step 1 comprises: Step 1.

1. Establish the kinematic model of the unmanned metamorphic vehicle based on the homogeneous coordinate transformation method; Construct any i-1 joint coordinate system among the ankle, knee, and hip joints in the leg mechanism ; The angle of clockwise rotation around the coordinate axis is positive, and the opposite is negative; ; Indicates the number of components, and n=6; Assume the i-1th joint coordinate system Around The pitch rotation angle of the axis is , then along Axis translation , and then along Axis translation Then, we get the i-th joint coordinate system ; Using formula (1) to get the i-th joint coordinate system Relative to the i-1th joint coordinate system The coordinate transformation matrix , thus obtaining the homogeneous transformation matrix of each joint coordinate system relative to the basic coordinate system in turn according to formula (1), including: the homogeneous transformation matrix of the right ankle joint coordinate system relative to the basic coordinate system , the homogeneous transformation matrix of the right knee joint coordinate system relative to the right ankle coordinate system , the homogeneous transformation matrix of the right hip joint coordinate system relative to the right knee joint coordinate system , the homogeneous transformation matrix of the left hip joint coordinate system relative to the right hip joint coordinate system , the homogeneous transformation matrix of the left knee joint coordinate system relative to the left hip joint coordinate system , the homogeneous transformation matrix of the left ankle joint coordinate system relative to the left knee joint coordinate system , where the origin of the base coordinate system is at the center of the supporting foot; (1) Formula (2) is used to obtain the homogeneous transformation matrix of each joint coordinate system relative to the basic coordinate system during walking, thereby obtaining the position coordinates of the center of mass of each component relative to the basic coordinate system during walking. Among them, the coordinate transformation matrix of the ankle joint of the swinging leg relative to the basic coordinate system during walking is : (2) Formula (3) is used to obtain the homogeneous transformation matrix of each joint coordinate system relative to the basic coordinate system during the reconstruction process, thereby obtaining the position coordinates of the center of mass of each component relative to the basic coordinate system during the reconstruction process. Among them, the coordinate transformation matrix of the hip joint coordinate system of any supporting leg relative to the basic coordinate system during the reconstruction process is ; (3) Step 1.

2. Establish a dynamic model of the unmanned metamorphic vehicle based on Lagrangian dynamics; According to the analytical mechanics theory, the Lagrangian dynamics model of the leg mechanism of the unmanned metamorphic vehicle during walking and reconstruction is established using formula (4): (4) In formula (4), is with The relevant generalized force, for The derivative of represents the integral, Indicates time, The kinetic energy of the unmanned transforming car, The potential energy of the unmanned transforming car, is the Lagrangian operator of the unmanned vehicle; and: (5) (6) In formula (5) and formula (6), for The derivative of express No. elements, represents transpose; represents the position coordinates of the center of mass of the i-th component in the basic coordinate system, represents the mass of the i-th component; Step 1.

3. Use the zero moment point theory ZMP as the stability criterion of the unmanned metamorphic vehicle and use formula (7) to establish the stability index of the unmanned metamorphic vehicle in the X-axis direction during the walking process and reconstruction motion. And the stability index of the Y axis direction ; (7) In formula (7), represents the second-order derivative of the Z-axis coordinate of the i-th component in the basic coordinate system, represents the second-order derivative of the X-axis coordinate of the i-th component in the basic coordinate system, Indicates the Z-axis coordinate of the i-th component in the basic coordinate system, represents the Y-axis coordinate of the i-th component in the basic coordinate system, Represents the acceleration due to gravity.

3. The structural parameter optimization method of the leg mechanism of an unmanned metamorphic vehicle according to claim 2 is characterized in that: The step 2 includes: Step 2.1: Set the constraints of the optimization variables; Formula (8) is used to establish the constraints of the unmanned metamorphic vehicle under reconstruction motion: (8) In formula (8), and Represent the angles of the ankle and knee joints at the start of reconstruction, Indicates the length of the sole of the foot. Indicates the distance between the leg mechanism and the front extreme position of the vehicle body during the deployment process under the reconstruction movement, Indicates the height from the ground to the center of rotation of the ankle joint around the Y axis in its own joint coordinate system. Indicates the height between the rotation center of the Y-axis actuator of the hip joint and the mounting plate of the leg mechanism. Indicates the height between the mounting plate of the leg mechanism and the horizontal ground in the vehicle state; Formula (9) is used to establish the constraints of the unmanned vehicle during its movement: (9) In formula (9), represents the center of mass height planned by the inverted pendulum model during walking; Formula (10) is used to establish the leg length ratio constraint of the unmanned metamorphic vehicle: (10) In formula (10), Indicates the minimum ratio of thigh length to calf length of the leg mechanism; Indicates the maximum ratio of thigh length to calf length of the leg mechanism; Step 2.2: Establish the objective function of the optimization variables, including: motion energy consumption function and motion stability function; Formula (11) is used to establish the power consumed by the i-th joint of the unmanned metamorphic vehicle during walking or reconstruction motion: : (11) Formula (12) is used to establish the total energy consumption objective function of all joints of the unmanned metamorphic vehicle during walking. And the total energy consumption objective function of all joints in the reconstruction movement : (12) In formula (12), represents the energy consumed by the i-th joint during walking or reconstruction movement; Indicates the total duration of the walking process. represents the total duration of the motion during the reconstruction process; Let the expected zero moment point ZMP in the X direction during the motion of the unmanned vehicle be x,d =0, the desired zero moment point in the Y direction is ZMP y,d ; Thus, the stability objective function of the unmanned metamorphic vehicle during the walking process is determined using formula (13) and the stability objective function of all joints in the reconstructed motion ; (13) In formula (13), Indicates the true zero moment point in the X direction; Indicates the true zero moment point in the Y direction; Using formula (14), we can get the single objective function in the walking process and the reconstruction motion: : (14) In formula (14), 、 、 、 Represents 4 weights.

4. The structural parameter optimization method of the leg mechanism of an unmanned metamorphic vehicle according to claim 3 is characterized in that: The step 3 includes: Step 3.0: Define the current number of iterations as g and initialize g=1; the maximum number of iterations is ; Let the total number of ants in the ant population be m, and each ant individual represents a set of optimization variables; Step 3.1: Initialize the ant number k=1; Define that any k-th ant in the g-th generation ant population contains M genes. The first M / 2 genes represent the leg length of the thigh, and the last M / 2 genes represent the leg length of the calf. Denotes the s-th gene of any k-th ant in the g-th generation ant population, let The value of ,Depend on The sth node coordinates of any kth ant in the gth generation ant population; Randomly initialize the pheromone concentration on the path connecting the sth node coordinate and the s+1th node coordinate shared by all ants in the gth generation ant population ; Step 3.2: Initialize s=1; in the feasible interval [ , ] randomly initialize the value of each gene of the k-th ant in the g-th generation ant population, and form the s-th node coordinate of the k-th ant in the g-th generation ant population; randomly select an ant in the g-th generation ant population as the optimal ant, and let the value of its s-th gene be ; Step 3.3: Use formula (15) to calculate the visibility information of any k-th ant in the g-th generation ant population on the connection path between the s-th node coordinate and the s+1-th node coordinate : (15) In formula (15), represents the number of any k-th ant in the g-th generation ant population. The value of the bit gene, Indicates the best ant in the g-1 generation. The value of the bit gene, ( ) represents the set of all next feasible node coordinates of the sth node coordinate of the kth ant in the gth iteration ( ) in the feasible node coordinates; Step 3.4: Use formula (16) to calculate the distance that any k-th ant in the g-th generation ant population crawls from the s-th node coordinate to ( ) in the state transition probability of the s+1th node coordinate , thus according to , take the gene value corresponding to the selected s+1th node coordinate As the kth ant in the gth generation of ants, alleles The updated value; (16) In formula (16), represents the pheromone concentration on the connection path between the sth node coordinate and the s+1th node coordinate shared by all ants in the gth iteration, is the visibility information on the connection path between the coordinates of the sth node and the coordinates of the s+1th node in the kth ant at the gth iteration, and They are pheromone factor and visibility factor respectively; Step 3.5: After assigning s+1 to s, if s>S, the updated gene value of each gene of the k-th ant in the g-th generation ant population is obtained, and step 3.5 is executed; otherwise, return to step 3.3 and execute sequentially; Step 3.6: After assigning k+1 to k, if k>m, the updated gene values ​​of each gene of all ants in the g-th generation ant population are obtained. The ant with the smallest objective function value in the g-th generation ant population is calculated as the optimal ant of the g-th generation, and step 3.7 is executed. Otherwise, return to step 3.2 and execute sequentially. Step 3.7: Determine whether the objective function value of the best ant of generation g is less than that of the best ant of generation g-1. If so, the best ant of generation g remains unchanged. Otherwise, the best ant of generation g-1 is assigned to the best ant of generation g. Step 3.8: Determine whether g = G. If so, stop the iteration and output the optimal ant of the Gth generation ant population as the optimal thigh length and optimal calf length. Otherwise, proceed to step 3.

9. Step 3.9: Use formula (17) to calculate the pheromone increment of any k-th ant in the g-th generation ant population crawling from the s-th node coordinate to the s+1-th node coordinate : (17) In formula (17), is the pheromone intensity, is the objective function of any k-th ant in the g-th generation ant population. When g=1, let ; Step 3.10: Use formula (18) to calculate the sum of pheromone increments on the connection path between the sth node coordinate and the s+1th node coordinate shared by all ants in the gth generation ant population. , thus calculating the The pheromone concentration shared by all ants in the generation ant population crawling from the sth node coordinate point to the s+1th node coordinate point : (18) In formula (18), is a constant that reflects the evaporation rate of pheromones, and <1; Step 3.11: After assigning g+1 to g, step 3.2 is executed sequentially.

5. An electronic device comprising a memory and a processor, characterized in that: The memory is used to store a program that supports a processor to execute the optimization method according to any one of claims 1 to 4, and the processor is configured to execute the program stored in the memory.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the optimization method according to any one of claims 1 to 4 are executed.

Images