Walking control system

Abstract

The objective is to provide a walking control system that can control the movement of a walking robot in a semi-passive manner. [Solution] A walking control system for controlling the walking speed of a semi-passive device having legs, wherein the model of the semi-passive device consists of a mass that is the main body, a leg link extending from the main body and equipped with a vibration damping mechanism, and a leg mass located on the leg link but not at the end of the link, and comprises means for supplying initial energy to the semi-passive device to start moving, and a control unit for changing the parameters of the vibration damping mechanism and controlling the walking speed of the semi-passive device.

Description

[Technical Field]

[0001] This invention relates to a walking control system. [Background technology]

[0002] A passive walker can walk stably, for example, down a slope, without the need for actuators. In a simple passive walker, two rigid legs are connected by the rotation of the hip joint. Such passive walking support legs perform a motion similar to an inverted pendulum. The SLIP (spring-loaded inverted pendulum) model is a model in which the support legs are spring elements and can reproduce human running. However, the SLIP model ignores the mass of the legs and assumes that the swinging legs make contact with the ground at a desired angle.

[0003] For example, the technology described in Patent Document 1 forced the robot's movement without utilizing passive dynamics. [Prior art documents] [Patent Documents]

[0004] [Patent Document 1] Japanese Patent Publication No. 2023-55542 [Overview of the project] [Problems that the invention aims to solve]

[0005] However, while conventional technology could perform semi-passive actions, it was not possible to precisely control walking, such as making the robot walk at a target speed.

[0006] The present invention has been made in view of the above-mentioned problems, and aims to provide a walking control system that can control the movement of a walking robot in a semi-passive manner. [Means for solving the problem]

[0007] (1) To achieve the above object, a walking control system according to one aspect of the present invention is a walking control system that controls the walking speed of a quasi-passive device having legs. The model of the quasi-passive device includes a mass point as the main body, a leg link extending from the main body and provided with a vibration damping mechanism, and a leg mass point located on the leg link at a position other than the link end. The walking control system further includes means for giving initial energy for starting movement to the quasi-passive device, and a control unit for changing parameters of the vibration damping mechanism to control the walking speed of the quasi-passive device.

[0008] (2) In the walking control system according to one aspect described in (1) above, the control unit may set the parameters of the vibration damping mechanism such that the eigenvalue of the Poincare map of the equation of motion of the model of the quasi-passive device is less than 1.

[0009] (3) In the walking control system according to one aspect described in (1) or (2) above, the control unit may increase the damping ratio, which is one of the parameters of the vibration damping mechanism, to decrease the walking speed.

[0010] (4) In the walking control system according to one aspect described in (1) or (2) above, the control unit may decrease the damping ratio, which is one of the parameters of the vibration damping mechanism, to increase the walking speed.

[0011] (5) In the walking control system according to one aspect described in any one of (1) to (4) above, walking includes a double-leg support phase in which the tips of the ground sides of the two legs are in contact with the ground, and a single-leg support phase in which the tip of the support leg, which is one of the two legs, is in contact with the ground and the tip of the floating leg, which is the remaining leg, is not in contact with the ground. The model of the quasi-passive device may further include a virtual mass at the tip of the ground side of the floating leg in the double-leg support phase.

[0012] (6) In the walking control system according to one aspect described in (5) above, assuming that each of the first force generated by the vibration damping mechanism provided in the support leg and the second force generated by the vibration damping mechanism provided in the floating leg is only a positive value in the extending direction of the leg, control may be performed by imposing a constraint condition that the first force is 0 when the first force is less than 0, and the second force is 0 when the second force is less than 0.

[0013] (7) In the walking control system according to one aspect described in (6) above, when the foot of the floating leg is in contact with the ground, control may be performed by imposing a holonomic constraint on the equation of motion of the model of the quasi-passive device.

[0014] (8) In the walking control system according to one aspect described in any one of (1) to (7) above, the vibration damping mechanism may include a spring and a damper, and a mechanism that can vary each of the parameters of the spring and the parameters of the damper.

Advantages of the Invention

[0015] According to the aspects (1) to (8) above, the operation of the walking robot can be controlled quasi-passively.

Brief Description of the Drawings

[0016] [Figure 1] It is a diagram showing a passive walking model having a viscoelastic element at the foot tip. [Figure 2] It is a diagram showing a model of a quasi-passive device with a mass mT added to the tip of the floating leg. [Figure 3] It is a diagram showing a configuration example of the walking control system of the embodiment. [Figure 4] It is a flowchart of the processing of the control device of the embodiment. [Figure 5] It is a diagram showing the values of the parameters used in the numerical simulation. [Figure 6] It is a diagram showing an example of the simulation result. [Figure 7]This figure shows an example of the results of a simulation of the relationship between walking speed and incline angle. [Figure 8] This figure shows an example of simulation results regarding the relationship between the slope angle, angular velocity, and the maximum absolute value of the eigenvalue. [Modes for carrying out the invention]

[0017] Embodiments of the present invention will be described below with reference to the drawings. Note that in the drawings used in the following description, the scale of each component has been appropriately changed to ensure that each component is recognizable. In all the figures used to illustrate the embodiments, components with the same function are given the same reference numerals, and repeated explanations are omitted. Furthermore, in this application, "based on XX" means "based on at least XX," and includes cases where it is based on another element in addition to XX. Also, "based on XX" is not limited to cases where XX is used directly, but also includes cases where it is based on something that has been calculated or processed from XX. "XX" is any element (for example, any information).

[0018] <Modeling of a passive walker> First, we will explain the modeling of a passive walker using viscoelastic elements at the tips of the legs. In this embodiment, for example, a passive walking machine is the subject. Passive walking is walking that does not have a power device, for example, by moving the left and right legs alternately by weight to descend a slope (see, for example, Reference 1).

[0019] Reference 1; Tetsuya Kinugasa, Koichi Osuga, et al., "Recommendation for Passive Walking Robots: How to Build a Bipedal Robot Using Only Gravity," Corona Publishing Co., Ltd., 2016.

[0020] Figure 1 shows a passive walking model with viscoelastic elements at the tips of the feet. The walking motion is limited to a downhill slope. In the following explanation, the state where both feet are in contact with the ground is referred to as the double-leg support stage (symbols g1, g2), and the state where only one foot is in contact with the ground is referred to as the single-leg support stage (symbols g3, g4). In the double-leg support stage, each leg has a viscoelastic element with a spring and damper arranged in parallel. Symbol g1 represents the model parameter in the single-leg support stage. Symbol g2 represents the dynamic variable in the single-leg support stage. Symbol g3 represents the model parameter in the double-leg support stage. Symbol g4 represents the dynamic variable in the double-leg support stage. It is assumed that the tips of the feet do not slip in this state. Therefore, there are two independent variables. In Figure 1, the viscoelastic element (vibration reduction mechanism) of the swing leg is omitted in symbols g1 and g2 because it does not affect walking.

[0021] In the walking model, the two legs are connected by a frictionless axis of rotation at the hip joint. In Figure 1, k is the spring constant, c is the damping coefficient, l1 is the length of the supporting leg, l2 is the length of the swing leg, m is the leg mass, and m Hθ is the hip joint mass. The leg mass m is located on the leg (leg link), not at the end of the leg. d is the distance from the hip joint to the mass m of each leg (leg mass). l0 is the leg length of the supporting leg when the spring is at its natural length, and l2 is equal to this when the leg is free (sign g2). In this model, the degrees of freedom in the single-leg support stage are reduced by placing the leg mass m between the hip joint and the viscoelastic element. In the single-leg support stage, the viscoelastic element of the free leg is fixed, and the free leg can be considered a rigid body. As a result, the degrees of freedom in the single-leg support stage are the angle of the supporting leg (θ1 in sign g2), the length of the supporting leg (l1 in sign g2), and the angle of the free leg (-θ2 in sign g2). The transition from the single-leg support stage to the double-leg support stage occurs when the tip of the free leg touches the ground. Note that sudden forces and collisions with the ground are ignored when the free leg swings forward. Furthermore, it is assumed that the contact point neither slides nor bounces during the collision between the swing leg and the ground. When the swing leg collides with the ground, the angular velocity and leg length velocity of the swing leg and the support leg change discontinuously due to the impact force from the ground. The angular velocity and leg length velocity immediately after the collision were obtained from the equation of motion for the single-support stage using Lagrange's collision equations.

[0022] (Equation of motion for single-leg support stage) The general equation of motion for a spring, mass, and damper system is: Mx ·· +Cx · It is given by +Kx=f. In this equation, M is the mass, K is the elastic modulus, C is the viscosity modulus, and x is the direction of motion. Furthermore, the equation of motion depends on the contact conditions. The equation of motion during the single-leg support stage, when derived using the Euler-Lagrange method, is given by equation (1). In equation (1), Q is given by equation (2), and q is given by equation (3). Note that q · is the first derivative, q ·· This is the second derivative. Also, the superscript T represents the transpose. Note that in equation (1), M(q)q ·· This is a term relating to mass, and C(q,q · )q · The term is related to angular velocity, and G(q) is related to gravity.

[0023]

number

[0024]

Number

[0025]

Number

[0026] The force f1 generated by the viscoelastic element at the foothold is given by the following equation (4).

[0027]

Number

[0028] Assume that the force due to the viscoelastic element acts only in the extension direction of the leg, i.e., only with positive values. Therefore, since the force due to the viscoelastic element does not act in the contraction direction, as a result of Equation (4), when f1 < 0, f 1= is assumed to be 0.

[0029] (Equation of motion for the double-support phase) In the double-support phase, as shown in Fig. 1, both legs include viscoelastic elements. Therefore, first, as shown in Fig. 2, the equation of motion for single-leg support with a mass m T at the tip of the free leg is derived. Fig. 2 is a diagram showing a model of a quasi-passive device (e.g., a robot) with a mass m T added to the tip of the free leg. It is assumed that the tip of the foot does not slip.

[0030] As shown in Fig. 2, the walking control system 1 controls the moving speed of a quasi-passive device (e.g., a passive walker 2) having legs (e.g., the first leg 21 (Fig. 3), the second leg 22 (Fig. 3)). Also, the model g5 of the quasi-passive device, as shown in Fig. 2, is "a body (body) that is a mass point (m HThe walking control system 1 consists of a "leg link with an extension damping mechanism from the main body" (g11) and a "leg mass point (m) located on the leg link but not at the end of the link." The walking control system 1 also includes means (vibration damping mechanism 212, vibration damping mechanism 222) (g12) to provide initial energy to the quasi-passive device when it starts moving. Furthermore, the walking control system 1 includes a control unit 31 that changes the parameters of the vibration damping mechanism and controls the movement speed of the quasi-passive device.

[0031] In this embodiment, the model g5 of the quasi-passive device shown in Figure 2 is stored in the memory unit 32 (Figure 3) of the control device 3, which includes the control unit 31. The control unit 31 controls the walking speed by changing the parameters of the vibration damping mechanism (spring and damper) by referring to this model g5 of the quasi-passive device. The vibration damping mechanism in this embodiment is a mechanism that allows the parameters of the spring and the damper to be varied.

[0032] Here, by applying a constraint force to the tip of the swing leg, the equation of motion for the double-leg support stage can be obtained. There are two independent variables at this stage. Mass m at the tip of the free leg T The equation of motion for a single-leg support with is given by equation (5). In equation (5), q d The equation is given by (6), and Q d This is given by equation (7). Note that in equation (5), M d (q d )q d ·· This is a term relating to mass, C d (q d ,q d · )q · This is a term relating to angular velocity, G d (q d ) is a term related to gravity.

[0033]

number

[0034]

number

[0035]

number

[0036] In equation (7), f1 is the force generated by the viscoelastic element of the stance leg and is expressed by equation (8). f2 is the force generated by the viscoelastic element of the swing leg and is expressed by equation (9).

[0037]

number

[0038]

number

[0039] We assume that the force due to the viscoelastic element is only positive, in the direction of leg extension. Therefore, in equation (8), when f1 < 0, f1 = 0. Also, in equation (9), when f2 < 0, f2 = 0. As a result, in this embodiment, the control unit 31 generates a virtual mass (m) at the ground-side tip of the free leg during the single-leg support stage. T Using a model of a quasi-passive device further equipped with ), the equation of motion for the single-leg support stage was determined. In this embodiment, the control unit 31 assumes that the first force (f1) generated by the vibration damping mechanism of the support leg and the second force (f2) generated by the vibration damping mechanism of the free leg are both positive values ​​in the direction of leg extension, and controls the device by imposing constraints such that the first force is 0 when the first force is less than 0, and the second force is 0 when the second force is less than 0.

[0040] Furthermore, the equations of motion during the double-leg support phase are derived by adding constraint forces to the equations of motion. These constraint forces are derived from the contact state between the foot and the ground. When the swing leg is in contact with the ground, an independent holonomic constraint is imposed, as shown in equation (10). A holonomic constraint is one in which the constraint conditions can be analytically expressed by equations that depend only on generalized coordinates and time.

[0041]

number

[0042] Thus, the equation of motion considering the contact forces acting on bipedal locomotion can be expressed as equation (11). In equation (11), H is given by equation (12) and E is given by equation (13).

[0043]

number

[0044]

number

[0045]

number

[0046] In equation (12), λ is an external force and is given by equation (14).

[0047]

number

[0048] (Impact equation) Here, for simplicity, we derive the impact equation assuming that no rebound or slip occurs due to the impact. This means that when the swing leg touches the ground during the single-leg support phase, the movement transitions to the double-leg support phase. The constraint condition after impact satisfies equation (15).

[0049]

number

[0050] Furthermore, equation (15) satisfies equation (16). Also, in equation (16), E1 is given by equation (17).

[0051]

number

[0052]

number

[0053] As a result, the equation of impact is expressed by the following equation (18).

[0054]

number

[0055] In equation (18), λ I This is an impulse originating from the impact. From equations (16) and (18), the velocity immediately after the impact can be calculated as shown in equation (19).

[0056]

number

[0057] It is assumed that the transition from the two-legged support stage to the one-legged support stage occurs when the spring of the stance leg (rear leg) reaches its natural length.

[0058] <Study of periodic gait and stability> Next, we will examine stability during periodic walking. First, walking motion was investigated using the Poincaré map. The Poincaré map is a method for reducing a continuous dynamical system to a discrete dynamical system. Furthermore, the Poincaré section was determined immediately after the transition from the two-legged support stage to the one-legged support stage. The Poincaré map is given by the following equation (20). Note that a suitable line is drawn in the state space so as to intersect with the trajectory. This is called the Poincaré section. For more information on the Poincaré map, Poincaré section, etc., see, for example, reference 1.

[0059]

number

[0060] In equation (20), q is the state of the Poincaré cross section, and is given by the following equation (21).

[0061]

number

[0062] Furthermore, in equation (20), i q p This is the state at the i-th step. i+1 q p This is the state at the (i+1)th step.

[0063] The fixed point of the Poincaré map exhibits periodic motion. Fixed point q * The following equation (22) is satisfied.

[0064]

number

[0065] Furthermore, we define g as shown in equation (23).

[0066]

number

[0067] As a result, the fixed point satisfies equation (24).

[0068]

number

[0069] The Newton-Raphson method is used to determine the fixed point. As a result, the selected fixed point q p1 The Jacobian matrix in equation (26) can be calculated numerically.

[0070]

number

[0071] When performing walking motion, it is important to determine whether the periodic motion is stable or unstable. If the periodic motion is unstable, the robot will fall over even if the initial perturbations or disturbances are small. The Poincaré map is expressed by the following equation (26).

[0072]

number

[0073] From equations (22) and (26), we obtain equation (27).

[0074]

number

[0075] In equation (27), J is the Jacobian matrix and is given by equation (28).

[0076]

number

[0077] Also, in equation (27), q ^ p is a fixed point q * It is a small perturbation from and is given by equation (29).

[0078]

number

[0079] Here, if all the eigenvalues ​​of the Jacobian matrix lie inside the unit circle, then all sufficiently small perturbations converge to zero, and the periodic motion becomes asymptotically stable.

[0080] <Walking control system> Figure 3 shows an example of the configuration of the walking control system of this embodiment. As shown in Figure 3, the walking control system 1 includes, for example, a passive walking machine 2 and a control device 3. The passive walking machine 2 comprises, for example, a first leg 21, a second leg 22, and a communication unit 23. The first leg 21 comprises, for example, a mechanism 211, a vibration damping mechanism 212, and a sensor 213. The second leg 22 comprises, for example, a mechanism 221, a vibration damping mechanism 222, and a sensor 223. The control device 3 includes, for example, a control unit 31, a storage unit 32, a communication unit 34, and an acquisition unit 35. The storage unit 32 includes, for example, a region for storing model 33.

[0081] (Passive walking device) The passive walking robot 2 is, for example, a bipedal robot. The passive walking robot 2 has a body (main body) attached to the upper part of the first leg 21 and the second leg 22.

[0082] The first leg portion 21 corresponds, for example, to the right or left foot of a human. The mechanism 211 is a link of the first leg 21. The vibration damping mechanism 212 includes the spring constant k (Figures 1 and 2) and damping coefficient c (Figures 1 and 2) described above. Sensor 213 is a sensor that detects the angle θ1 of the first leg portion 21 (Figure 1).

[0083] The second leg portion 22 corresponds, for example, to the right or left foot of a human other than the first leg portion 21. The mechanism 221 is a link for the second leg 22. The vibration damping mechanism 222 includes the spring constant k (Figures 1 and 2) and damping coefficient c (Figures 1 and 2) described above. Sensor 223 is a sensor that detects the angle θ2 of the second leg portion 22 (Figure 1).

[0084] The communication unit 23 acquires control instructions output by the control device 3. The communication unit 23 outputs the detected values ​​detected by sensors 213 and 223 to the control device 3.

[0085] (Control device) The control device 3 and the passive walking device 2 are connected to each other via a wired or wireless network NW. The control device 3 may also be located within the passive walking device 2. The control device 3 also has a power supply and other necessary components.

[0086] The control unit 31 uses the detected values ​​obtained from the passive walker 2 and the data stored in the memory unit 32 to change the parameters of the vibration damping mechanisms 212 and 222 of the passive walker 2, and generates control instructions to control the movement speed of the passive walker 2. The damping ratio and other details will be described later. The control unit 31 also provides initial energy to the passive walker 2 (quasi-passive device) when it starts moving. The means of providing initial energy to the passive walker 2 (quasi-passive device) when it starts moving may be, for example, changing the angle of the downhill slope, or the worker pushing the passive walker 2 when starting to walk.

[0087] The memory unit 32 stores algorithms, programs, formulas, thresholds, and data related to the passive walking machine 2 (mass m) used by the control unit 31 for control. H It stores information such as mass m, distance d, leg length l0, and parameters.

[0088] Model 33 stores Model g5 of the quasi-passive device shown in Figure 2. Note that Model 33 may be stored on a server connected via a network, or it may be located in the cloud.

[0089] The communication unit 34 outputs the control instruction generated by the control unit 31 to the passive walking device 2. The communication unit 34 acquires the detected value output by the passive walking device 2.

[0090] The acquisition unit 35 acquires, for example, data related to the passive walking machine 2 and the angle of the slope.

[0091] The configuration shown in Figure 3 and the functions described above are merely examples and are not limited to them. For example, the control device 3 may be included within the passive walking device 2.

[0092] <Processing Procedure> Next, an example of the processing procedure performed by the control device 3 will be described. Figure 4 is a flowchart of the processing performed by the control device in this embodiment.

[0093] (Step S1) The control unit 31 stores the model (parameter) g5 of the quasi-passive device in the storage unit 32.

[0094] (Step S2) The acquisition unit 35 of the control device 3 acquires the target speed.

[0095] (Step S3) The control device 3 applies energy for starting the operation, for example, kicking energy. Note that the energy for starting the operation may be, for example, the gradient of a downhill slope.

[0096] (Step S4) The control unit 31 refers to the model g5 of the quasi-passive device and uses the formulas stored in the memory unit 32 to change the parameters of the vibration damping mechanism according to the target speed. In order to decrease the target speed, the control unit 31 increases the damping ratio, which is one of the parameters of the vibration damping mechanism. Alternatively, the control unit 31 decreases the damping ratio, which is one of the parameters of the vibration damping mechanism, in order to increase the walking speed.

[0097] Furthermore, the control device 3 may determine the walking speed based on the detected values ​​from the sensors 213 and 223 of the passive walking machine 2, and may further change the parameters of the vibration damping mechanism according to the determined walking speed.

[0098] The control device 3 may also be configured to provide initial energy to the quasi-passive device when walking begins. Alternatively, the user may provide initial energy to the quasi-passive device when walking begins.

[0099] <Simulation> Next, we will explain an example of the simulation results. Figure 5 shows the parameter values ​​used in the numerical simulation. As shown in Figure 5, mass m T The simulation was performed with a weight of 5 kg, leg mass m of 1 kg m, distance d of 0.25 m, and length l0 of 0.5 m. γ is the inclination angle. To investigate the effects of the spring constant k and damping coefficient c of a viscoelastic element on walking, the natural angular frequency ω without damping and the damping ratio ζ were used in the following equations (30) and (31).

[0100]

number

[0101]

number

[0102] In the simulation, the natural angular frequency without attenuation and the attenuation ratio were determined for the case where both legs were stationary vertically and the movement was supported by one leg. Figure 6 shows an example of the simulation results. The simulation conditions were periodic walking with a slope angle γ of 0.03 (rad), a natural angular frequency ω of 40 (rad / s), and an attenuation ratio ζ = 0.1. The walking speed was 0.482 m / s, and the maximum absolute value of the eigenvalues ​​of the Jacobian matrix was 0.77, indicating that the periodic walking motion is stable, as shown in Figure 6.

[0103] The graph labeled g10 shows the change in leg angle during periodic walking. The horizontal axis represents time (seconds), and the vertical axis represents angle (radians). Line g11 represents the angle θ1 of the support leg, and line g12 represents the angle θ2 of the swing leg. As indicated by symbol g10, the swing leg and the supporting leg switch places during the transition from the double-legged support stage to the single-legged support stage. The simulation results showed that the angle of the supporting leg increased monotonically, and the swinging leg swung backward at an angle of 0.35 rads and forward at an angle of -0.39 rads. When the swinging leg touched the ground, the angle of the swinging leg was -0.30 rads and the angle of the supporting leg was 0.27 rads. Furthermore, when the hind leg left the ground, the angle θ2 of the foreleg was -0.25 rads and the angle of the hind leg was 0.33 rads. The step time was 0.58 seconds.

[0104] The graph labeled g20 shows the change in length of the support leg and the free leg. The horizontal axis represents time (seconds), and the vertical axis represents the leg length (m). Line g21 represents the length l1 of the first leg 21, and line g22 represents the length l2 of the second leg 22. Furthermore, in the single-leg support stage, when the swing leg is not in contact with the ground, the length of the swing leg is assumed to be the length of the leg when the spring is at its natural length (l0). As mentioned above, the length of the leg when the spring is at its natural length is 0.5m. As shown in the graph with sign g20, the swing leg collides with the ground at 0.55 seconds, and the leg begins to contract. As the swing leg collides with the ground, the length of the supporting leg increases, and when the spring reaches its natural length of 0.5m, the foot leaves the ground.

[0105] The graph labeled g30 shows the change in force generated by the viscoelastic elements of the support leg and the free leg. The horizontal axis represents time (seconds), and the vertical axis represents force (N). Line g31 represents the force f1 generated by the support leg, and line g22 represents the force f2 generated by the free leg. The force generated by each leg is approximately proportional to the amount of spring compression, indicating that the spring force is dominant over the damping force.

[0106] The graph labeled g40 shows the trajectory of the hip joint when the tip of the supporting leg is at the starting point. The horizontal axis is X (m), and the vertical axis is Z (m). Line g41 is the trajectory of the hip joint in this embodiment with a viscoelastic element, and line g42 is the trajectory of the hip joint in a rigid configuration without a viscoelastic element (l1=l0). The trajectory of the hip joint when using viscoelastic legs shows minimal vibration and damping due to the spring.

[0107] (Relationship between slope angle γ and pace when the maximum value of the eigenvalue is 1 or less) Next, we will explain an example of the simulation results regarding the relationship between the slope angle γ and the step rate when the maximum value of the eigenvalues ​​(eigenvalues ​​of the Poincaré map) is 1 or less.

[0108] Figure 7 shows an example of simulation results for the relationship between walking speed and incline angle. It shows the relationship between incline angle γ and stride length when the damping ratio ζ is changed from 0.1 to 1 while keeping the undamped natural angular frequency ω (also simply called angular velocity) constant. Figure 8 shows an example of simulation results for the relationship between incline angle, angular velocity, and the maximum value of the absolute value of the eigenvalue.

[0109] In each graph in Figure 7, the horizontal axis represents the incline angle (rad), and the vertical axis represents the average walking speed (m / s). The graph labeled g100 shows the simulation results when the angular velocity ω is 40 (rad / s), the graph labeled g110 shows the results when the angular velocity ω is 60 (rad / s), the graph labeled g120 shows the results when the angular velocity ω is 80 (rad / s), and the graph labeled g130 shows the results when the angular velocity ω is 100 (rad / s).

[0110] In each graph in Figure 8, the horizontal axis represents the slope angle (rad), and the vertical axis represents the maximum absolute value of the eigenvalue. The graph with sign g200 shows the simulation results when the angular velocity ω is 40 (rad / s), the graph with sign g210 shows the results when the angular velocity ω is 60 (rad / s), the graph with sign g220 shows the results when the angular velocity ω is 80 (rad / s), and the graph with sign g230 shows the results when the angular velocity ω is 100 (rad / s).

[0111] In the graphs labeled g100 to g130 in Figure 7, lines g101, g111, g121, and g131 represent the results when the leg portion is a rigid body without viscoelastic elements. As shown in Figure 7, for all angular velocities ω and damping ratios ζ, in this embodiment where the leg portion is equipped with a viscoelastic element, the average walking speed increases as the inclination angle increases, similar to the case where the leg portion is a rigid body. Furthermore, as shown in Figure 7, when the incline angle and damping ratio ζ are the same, the average walking speed increases as the damping ratio ζ decreases. Furthermore, as shown in Figure 7, for any angular velocity ω, the average walking speed is closest to the walking speed of the rigid leg when the damping ratio ζ is 0.2, and greater than the walking speed of the rigid leg when the damping ratio ζ is 0.1. Furthermore, as shown in Figure 7, the change in average walking speed when the damping ratio ζ is changed is greater as the angular velocity ω decreases.

[0112] For example, when the inclination angle γ is 0.03 (rad), the average walking speed of the rigid leg is 0.455 (m / s). In contrast, in this embodiment, where the leg portion is equipped with a viscoelastic element, when the angular velocity ω is 40 (rad / s), the walking speed is 0.482 (m / s) when the damping ratio ζ is 0.1, 0.455 (m / s) when the damping ratio ζ is 0.2, and 0.213 (m / s) when the damping ratio ζ is 1.0.

[0113] Furthermore, in this embodiment, where the leg portion is equipped with a viscoelastic element, the walking speed is 1.06 times that of a rigid leg when the damping ratio ζ is 0.1, and 0.47 times when the damping ratio ζ is 1.0. On the other hand, when the angular velocity ω is 100 (rad / s), the walking speed is 0.464 (m / s) when the damping ratio ζ is 0.1, 0.451 (m / s) when the damping ratio ζ is 0.2, and 0.426 (m / s) when the damping ratio ζ is 1.0. Thus, in this embodiment, where the leg portion is equipped with a viscoelastic element, the walking speed is 1.02 times that of a rigid leg when the damping ratio ζ is 0.1, and 0.94 times that of a rigid leg when the damping ratio ζ is 1.0.

[0114] As shown above, the simulation revealed that the average walking speed remains almost unchanged even when the damping ratio ζ changes. Therefore, when the vibration frequency is small, that is, when the spring constant is small, the average walking speed can be significantly changed by changing the damping ratio ζ, or damping coefficient (parameter). For this reason, in this embodiment, the movement speed of the passive walking machine 2 is controlled by controlling the magnitude of the damping ratio ζ, which is one of the parameters of the vibration damping mechanism.

[0115] Here, we will explain an example of parameter setting for a vibration damping mechanism. If you want to walk efficiently, for example, in the graph labeled g100 in Figure 7, the left side with a smaller inclination angle is desirable. As mentioned above, the inclination angle is the initial energy at the start of walking. Therefore, in order to reduce energy and improve efficiency, it is preferable for this inclination angle to be small as well. However, when setting the parameters of the vibration damping mechanism to obtain the selected damping ratio, if the desired walking speed cannot be stably maintained, the inclination angle is increased to control the walking speed to the desired speed, and then the parameters of the vibration damping mechanism are set. Conversely, if the walking speed is too fast compared to the desired speed, the inclination angle is reduced. In this way, the control unit 31 sets the parameters of the vibration damping mechanism so that the inclination angle is small and the initial energy is as small as possible in order to walk efficiently.

[0116] Based on the above, in this embodiment, the Jacobian is obtained from the Poincaré map, eigenvalues ​​are found, and the obtained eigenvalues ​​are set to be 1 or less. Furthermore, in this embodiment, in addition to this condition, walking speed is controlled by changing the damping ratio in equation (31). Note that changing the damping ratio is done by changing the mass m in equation (31). H Since the mass m is a fixed value, the parameters of the damping coefficient c and the spring constant k are changed.

[0117] In this embodiment, a quasi-passive device model (walking model) is used, which has a vibration damping mechanism, which is a viscoelastic element, at the ground-side tip of the leg. In this embodiment, the walking speed is controlled by controlling the stiffness and damping of the leg by changing the damping ratio in equation (31), for example, as a parameter of this vibration damping mechanism. The natural frequency of one leg is used for stiffness, and the damping ratio is used for damping. Thus, in this embodiment, the walking characteristics are controlled by changing the stiffness and damping, which are parameters of the vibration damping mechanism. For example, increasing the stiffness results in walking closer to the walking model, and decreasing the stiffness allows for a greater change in walking characteristics due to changes in damping.

[0118] Furthermore, in this embodiment, in the walking model of Figure 2, a virtual mass m is attached to the ground-side tip of the swing leg during the single-leg support stage. T I decided to add it.

[0119] As a result, according to this embodiment, the operation of the passive walking machine 2 can be appropriately controlled in a semi-passive manner simply by adjusting the viscoelasticity of the toe. Furthermore, according to this embodiment, the passive walking machine 2, which is equipped with a sticky elastic leg and leg mass based on the semi-passive device model 33, can perform periodic movements and walk stably.

[0120] Furthermore, the configuration and control method of the passive walking machine 2 described above can be applied to, for example, bipedal robot systems, walking assist systems, and the like. Furthermore, although the above example described an example with two legs, it is not limited to this. The passive walking machine 2 may, for example, have two legs on each side, for a total of four legs.

[0121] Furthermore, a program to implement all or part of the functions of the control device 3 in this invention may be recorded on a computer-readable recording medium, and all or part of the processing performed by the control device 3 may be performed by loading the program recorded on this recording medium into a computer system and executing it. Herein, "computer system" includes hardware such as an OS and peripheral devices. Furthermore, "computer system" also includes a WWW system equipped with a homepage provisioning environment (or display environment). Furthermore, "computer-readable recording medium" refers to portable media such as flexible disks, magneto-optical disks, ROMs, CD-ROMs, and storage devices such as hard disks built into a computer system. Moreover, "computer-readable recording medium" also includes volatile memory (RAM) inside a computer system that acts as a server or client when a program is transmitted via a network such as the Internet or a communication line such as a telephone line, which holds the program for a certain period of time. Alternatively, some or all of these components may be implemented by hardware (including circuitry) such as LSI (Large Scale Integration), ASIC (Application Specific Integrated Circuit), FPGA (Field-Programmable Gate Array), GPU (Graphics Processing Unit), or SOC (System On Chip), or by the collaboration of software and hardware.

[0122] Furthermore, the above program may be transmitted from a computer system that stores the program in a memory device or the like to another computer system via a transmission medium or by transmission waves within the transmission medium. Here, the "transmission medium" for transmitting the program refers to a medium that has the function of transmitting information, such as a network (communication network) such as the Internet or a communication line (communication line) such as a telephone line. In addition, the above program may be for the purpose of realizing a part of the functions described above. Furthermore, it may be a so-called differential file (differential program) that can realize the functions described above in combination with a program already recorded in the computer system.

[0123] Although embodiments for carrying out the present invention have been described above using examples, the present invention is not limited in any way to these embodiments, and various modifications and substitutions can be made without departing from the spirit of the present invention. [Explanation of Symbols]

[0124] 1...Walking control system, 2...Passive walking machine, 3...Control device, 21...First leg, 22...Second leg, 23...Communication unit, 211,221...Mechanism unit, 212,222...Vibration damping mechanism, 213,223...Sensor, 31...Control unit, 32...Storage unit, 34...Communication unit, 35...Acquisition unit, 33...Model

Claims

1. A walking control system for controlling the walking speed of a semi-passive device having legs, The model of the aforementioned quasi-passive device is, It consists of a main body (a point mass), leg links extending from the main body and equipped with a vibration damping mechanism, and leg point masses located on the leg links but not at the link ends. Means for providing initial energy to start the movement of the quasi-passive device, A control unit that changes the parameters of the vibration damping mechanism and controls the walking speed of the quasi-passive device, A walking control system equipped with the following features.

2. The control unit, The parameters of the vibration damping mechanism are set such that the eigenvalues ​​of the Poincaré map of the equations of motion of the quasi-passive device model are less than 1. The walking control system according to claim 1.

3. The control unit, In order to reduce the walking speed, the damping ratio, which is one of the parameters of the vibration damping mechanism, is increased. A walking control system according to claim 1 or claim 2.

4. The control unit, In order to increase the walking speed, the damping ratio, which is one of the parameters of the vibration damping mechanism, is reduced. A walking control system according to claim 1 or claim 2.

5. Walking comprises a two-legged support stage in which the ground-side tips of both legs are in contact with the ground, and a one-legged support stage in which the tip of one of the two legs, the support leg, is in contact with the ground, while the tip of the remaining leg, the swing leg, is not in contact with the ground. The aforementioned quasi-passive device model further includes a virtual mass at the ground-side tip of the swing leg during the bilateral leg support stage. A walking control system according to claim 1 or claim 2.

6. The control unit, It is assumed that the first force generated by the vibration damping mechanism of the support leg and the second force generated by the vibration damping mechanism of the free leg are both positive values ​​in the direction of extension of the leg, and the control is performed by imposing constraints such that the first force is 0 when the first force is less than 0, and the second force is 0 when the second force is less than 0. The walking control system according to claim 5.

7. When the foot of the swing leg is in contact with the ground, a holonomic constraint is imposed on the equation of motion of the quasi-passive device model to control it. The walking control system according to claim 6.

8. The vibration damping mechanism comprises a spring and a damper, This mechanism allows for the variable control of both the spring parameter and the damper parameter. A walking control system according to claim 1 or claim 2.

Images