Operation method of front elbow and rear knee two-stage motor-driven quadruped robot

By using a two-stage motor system (front elbow and rear knee) to drive the quadruped robot, differentiated gait planning and real-time collaborative control were achieved. This solved the problems of drive coordination and terrain adaptability of quadruped robots in the 'front elbow and rear knee' configuration, improving stability and flexibility, adapting to complex terrain, and extending equipment life.

CN121106532APending Publication Date: 2025-12-12CHENGDU JINFA EDGE INTELLIGENT TECHNOLOGY CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202511421741.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-09-30
Publication Date
2025-12-12

AI Technical Summary

Technical Problem

Existing quadruped robots in the 'front elbow, rear knee' configuration lack sufficient drive coordination, gait adaptability, and terrain adaptability, making it difficult to meet the requirements for stable and flexible movement in complex scenarios.

Method used

It adopts a two-stage motor drive scheme with front elbow and rear knee. Through differentiated gait planning and real-time collaborative control, the first-stage motors in the elbow and knee provide basic power, while the second-stage motors perform fine adjustments. Combined with real-time feedback from sensors, the motor parameters are dynamically adjusted to adapt to different gait and terrain.

Benefits of technology

It achieves a balance between stability and flexibility in complex terrain, improves motion efficiency and terrain adaptability, reduces the pressure on motor control precision and energy loss, and extends the service life of the equipment.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN121106532A_ABST
    Figure CN121106532A_ABST
Patent Text Reader

Abstract

The invention discloses an operation method of a front elbow and rear knee two-stage motor-driven quadruped robot, the quadruped robot comprises a trunk, a left forelimb, a right forelimb and other structures, the left forelimb and the right forelimb are arranged on the front portion of the trunk and are both of an elbow type structure, and the operation method of the quadruped robot comprises the steps that S1, according to the preset operation requirement of the quadruped robot, the left forelimb and the right forelimb are connected; determining the time proportions of the supporting phases and the swinging phases of the left foreleg, the right foreleg, the left posterior limb and the right posterior limb in one gait cycle. According to different preset operation requirements of low-speed stability, medium-speed high efficiency and the like, accurate matching is realized through differentiated gait planning, a walking gait is adopted for a low-speed scene, and stable movement under a complex terrain is guaranteed; diagonal gaits are adopted in a medium-speed scene, and the exercise efficiency is improved. And meanwhile, matched driving force and adjustment precision are provided for different gaits through cooperative control of the two-stage motor, so that the quadruped robot can flexibly cope with multiple scenes from daily inspection to medium-speed operation.
Need to check novelty before this filing date? Find Prior Art

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of robot movement, in particular to a running method of a four-legged robot driven by a front-elbow and rear-knee two-stage motor. BACKGROUND

[0002] At present, four-legged robots have broad application prospects in complex scenes such as rescue and disaster relief, field exploration, and special task execution due to their excellent terrain adaptability, and have become a research hotspot in the field of robots. With the continuous upgrading of application requirements, higher requirements are put forward for the motion stability, gait flexibility and complex terrain adaptation ability of four-legged robots.

[0003] At present, most four-legged robots adopt symmetrical limb structure or single-stage motor driving scheme, which has obvious technical limitations. On the one hand, the symmetrical structure is difficult to match the efficient motion configuration of four-legged animals (such as dogs) in the biological world, i.e. the structure of the biological forelimb elbow joint facing forward and the hindlimb knee joint facing backward, which can form a stable mechanical triangle during support, buffer ground impact and optimize load distribution. However, the existing symmetrical structure cannot fully utilize this biomechanical advantage, resulting in difficulty in balancing motion efficiency and stability. On the other hand, single-stage motor driving can only realize basic motion control of the joint, and cannot meet the dual needs of "large torque support" and "fine angle adjustment": it is difficult to stably support the trunk load during the support phase, and it is difficult to accurately plan the limb end trajectory during the swing phase, especially when facing rugged terrain, which easily causes trunk jolting, foot deviation and other problems.

[0004] For a small number of four-legged robots adopting the "front-elbow and rear-knee" special configuration, the existing technology has not yet broken through the bottleneck of insufficient driving coordination. The motion logic of the forelimb elbow joint and the hindlimb knee joint of such robots is significantly different, and needs to match different driving torques and angle adjustment rules, but the existing driving scheme lacks time sequence planning and gait adaptation strategy for two-stage motors (one stage is responsible for basic flexion and extension, and the second stage is responsible for fine swing). For example, when switching between different gaits such as walking, trotting and running, it is difficult to dynamically adjust the speed and torque parameters of the two-stage motor; when facing uneven terrain such as bumps and depressions, it is difficult to adjust the joint angle in real time through the second-stage motor to adapt to terrain changes; during steering, it is also difficult to achieve smooth steering through differential control and trajectory optimization of the motors on both sides of the limbs.

[0005] In summary, the existing four-legged robot technology has obvious shortcomings in the driving coordination, gait adaptability and terrain adaptability of the "front-elbow and rear-knee" configuration, and cannot meet the needs of stable and flexible motion in complex scenes. SUMMARY

[0006] The present application aims to overcome the defects of poor driving coordination, poor gait adaptability and poor terrain adaptability of the current quadruped robot in the "front elbow and rear knee" configuration, and to provide an operation method of a front elbow and rear knee two-stage motor driven quadruped robot which can not only fully release the motion potential of the special configuration, but also meet the stable and flexible motion demand in complex scenes.

[0007] The present application aims to overcome the defects of poor driving coordination, poor gait adaptability and poor terrain adaptability of the current quadruped robot in the "front elbow and rear knee" configuration, and to provide an operation method of a front elbow and rear knee two-stage motor driven quadruped robot which can not only fully release the motion potential of the special configuration, but also meet the stable and flexible motion demand in complex scenes. S1, according to the preset operation requirement of the quadruped robot, determining the time proportion of the support phase and the swing phase of the left front limb, the right front limb, the left rear limb and the right rear limb in a gait cycle, and the gait phase difference between any two limbs; S2, when the quadruped robot is in the support phase, the elbow primary drive motor and the knee primary drive motor output corresponding torque to support the trunk, and at the same time, the elbow secondary drive motor and the knee secondary drive motor fine-tune the joint angle to adapt to the ground state; when the quadruped robot is in the swing phase, the elbow primary drive motor and the knee primary drive motor drive the joint flexion to drive the limb swing, and at the same time, the elbow secondary drive motor and the knee secondary drive motor adjust the joint swing angle to plan the predetermined swing trajectory of the limb end; S3, real-time acquisition of the running state data of the quadruped robot, and dynamic adjustment of the control parameters of the elbow primary drive motor, the elbow secondary drive motor, the knee primary drive motor and the knee secondary drive motor according to the running state data, to ensure that the trunk posture of the quadruped robot is in a stable state.

[0008] Further, in step S1, when the preset operation requirement is medium-speed and high-efficiency operation, the diagonal gait planning is adopted, the left front limb and the right rear limb are divided into a first limb group, and the right front limb and the left rear limb are divided into a second limb group; the gait phase difference of the first limb group and the second limb group is 180°, and the support phase time proportion and the swing phase time proportion of a single limb are both 50%.

[0009] When the preset operation requirement in step S1 is low-speed and stable operation, the walking gait planning is adopted, and the left front limb, the right rear limb, the right front limb and the left rear limb enter the support phase in turn, the gait phase difference of the adjacent two limbs is 90°, and the support phase time proportion of a single limb is 60%, and the swing phase time proportion is 40%.

[0010] The four-legged robot in the support phase in step S2 includes an initial support phase and a middle support phase; in the initial support phase, when the elbow primary drive motor of the left front limb and the right front limb drives the elbow to extend, the knee primary drive motor of the left rear limb and the right rear limb drives the knee to bend, and the trunk load is jointly borne; in the middle support phase, when the elbow secondary drive motor of the left front limb and the right front limb adjusts the elbow swing angle, the knee secondary drive motor of the left rear limb and the right rear limb synchronously adjusts the knee swing angle, and the trunk horizontal posture is maintained.

[0011] The four-legged robot in the swing phase in step S2 includes an initial swing phase and a middle swing phase; in the initial swing phase, when the elbow primary drive motor of the left front limb and the right front limb drives the elbow to bend, the knee primary drive motor of the left rear limb and the right rear limb drives the knee to extend, and the limb is separated from the ground; in the middle swing phase, when the elbow secondary drive motor of the left front limb and the right front limb controls the limb end height from the ground in the interval of 10-15 cm, the knee secondary drive motor of the left rear limb and the right rear limb controls the limb end height from the ground in the interval of 8-12 cm, and the limb end swings to the landing position according to the predetermined trajectory.

[0012] The specific manner of the control parameter dynamic adjustment of the elbow primary drive motor, the elbow secondary drive motor, the knee primary drive motor and the knee secondary drive motor according to the running state data in step S3 is as follows: if the trunk pitch angle fluctuation exceeds 1°, the output torque of the elbow primary drive motor of the left front limb and the right front limb in the support phase is increased or the extension speed of the knee primary drive motor of the left rear limb and the right rear limb in the swing phase is adjusted; if the contact pressure of any limb of the left front limb, the right front limb, the left rear limb and the right rear limb exceeds the preset threshold, the swing angle of the corresponding secondary drive motor is finely adjusted to disperse the trunk load.

[0013] As one of the preferred manners, the running method further includes terrain adaptability adjustment; when the four-legged robot detects that there is a protruding or recessed terrain in front, if the left front limb or the right front limb contacts the protruding terrain, the elbow secondary drive motor of the left front limb or the right front limb swings to the protruding side, and the extension speed of the elbow primary drive motor of the left front limb or the right front limb is simultaneously reduced; if the left rear limb or the right rear limb contacts the recessed terrain, the knee secondary drive motor of the left rear limb or the right rear limb swings to the recessed side, and the bending speed of the knee primary drive motor of the left rear limb or the right rear limb is simultaneously reduced.

[0014] As a preferred mode two, the operation method further comprises a steering control, when the four-legged robot turns left, the elbow secondary drive motor of the left front limb and the knee secondary drive motor of the left rear limb increase the inside swing, the elbow primary drive motor of the right front limb and the knee primary drive motor of the right rear limb all increase the driving speed to form the motion differential; when the four-legged robot turns right, the elbow secondary drive motor of the right front limb and the knee secondary drive motor of the right rear limb increase the inside swing, the elbow primary drive motor of the left front limb and the knee primary drive motor of the left rear limb all increase the driving speed to form the motion differential.

[0015] Compared with the prior art, the present application has the following advantages and beneficial effects: (1) The present application precisely matches different preset operation requirements such as low-speed stability and medium-speed efficiency through differential gait planning. The walking gait is adopted for low-speed scenes to ensure smooth movement on complex terrain, and the diagonal gait is adopted for medium-speed scenes to improve motion efficiency. The cooperative control of the two-stage motors provides adaptive driving force and adjustment accuracy for different gaits, so that the four-legged robot can flexibly cope with multiple scenes from daily inspection to medium-speed operation.

[0016] (2) In the support phase, the primary motor provides stable support torque to bear the torso load, and the secondary motor synchronously fine-tunes the joint angle to adapt to the ground state, avoiding torso tilting caused by slight ground undulations. In the swing phase, the primary motor drives the limb to efficiently flex and extend, and the secondary motor precisely controls the limb end height and swing trajectory to ensure accurate foot landing. With real-time feedback adjustment, the torso bouncing amplitude and attitude fluctuation can be further controlled within a very small range, ensuring stability during operation.

[0017] (3) The present application can adapt to uneven terrain such as protrusions and depressions through early prediction and active adjustment of the secondary motor, and this adjustment logic does not rely on complex terrain preprocessing, which can greatly improve the passing ability of the robot in complex terrain scenes such as field exploration and rescue.

[0018] (4) The steering control of the present application adopts a double strategy of "differential + trajectory optimization". The steering side increases the inside swing through the elbow secondary drive motor and the knee secondary drive motor to shorten the propulsion distance, and the non-steering side increases the driving speed through the elbow primary drive motor and the knee primary drive motor to increase the propulsion force, forming a stable differential effect. This method does not require additional steering mechanisms, and the minimum steering radius is controllable, and the torso yaw angle fluctuation during steering is ≤2°, balancing flexibility and attitude stability, meeting the scene requirements of dynamic obstacle avoidance and path adjustment.

[0019] (5) The two-stage motors of this invention have a clear division of labor and close cooperation. The primary motors at the elbow and knee focus on "basic drive," undertaking the main power output for support and swing, ensuring motion efficiency. The secondary motors at the elbow and knee focus on "fine adjustment," responsible for angle correction and trajectory optimization, which can effectively reduce the control precision pressure on the primary motors at the elbow and knee and reduce energy loss. At the same time, through limb pressure feedback adjustment, the load of each limb is evenly distributed, avoiding motor losses caused by local overload, extending the service life of the equipment, and improving the reliability of long-term operation. Attached Figure Description

[0020] Figure 1 This is a schematic diagram of the overall structure of the quadruped robot of the present invention.

[0021] Figure 2 A schematic diagram of the overall process of the quadruped robot in operation according to the present invention.

[0022] The reference numerals in the above figures are named as follows: 1-Torso, 21-Elbow primary drive motor, 22-Elbow secondary drive motor, 31-Knee primary drive motor, 32-Knee secondary drive motor. Detailed Implementation

[0023] The present invention will be further described in detail below with reference to embodiments, but the implementation of the present invention is not limited thereto.

[0024] Example

[0025] like Figure 1 As shown, the quadruped robot of this embodiment includes a torso 1, a left forelimb and a right forelimb with elbow-type structures located on the left and right sides of the front part of the torso 1, and a left hindlimb and a right hindlimb with knee-type structures located on the left and right sides of the rear part of the torso 1. Simultaneously, each of the left and right forelimbs has an elbow primary drive motor 21 and an elbow secondary drive motor 22 inside, and each of the left and right hindlimbs has a knee primary drive motor 31 and a knee secondary drive motor 32 inside. That is, the left forelimb includes an elbow primary drive motor 21 and an elbow secondary drive motor 22, the right forelimb includes an elbow primary drive motor 21 and an elbow secondary drive motor 22, the left hindlimb includes a knee primary drive motor 31 and a knee secondary drive motor 32, and the right hindlimb includes a knee primary drive motor 31 and a knee secondary drive motor 32.

[0026] The elbow primary drive motor 21 is used to drive the basic flexion and extension movements of the elbows of the left and right forelimbs, providing basic driving force for support or swinging; the elbow secondary drive motor 22 is used to drive the swinging movements of the elbows of the left and right forelimbs, realizing fine adjustment of the elbow joint angle and adaptation of the limb trajectory.

[0027] The knee primary driving motor 31 is used to drive the flexion and extension basic action of the left and right hind limbs, and provide the basic driving force for supporting or swinging; the knee secondary driving motor 32 is used to drive the swing action of the left and right hind limbs, and realize the fine adjustment of the knee joint angle and the adaptation of the limb end trajectory.

[0028] For the convenience of subsequent description, the limbs in the embodiment are collectively referred to as the left front limb, the right front limb, the left hind limb and the right hind limb, and a single limb is a simple name of any one of the left front limb, the right front limb, the left hind limb or the right hind limb. The front limb is collectively referred to as the left front limb and the right front limb, and the hind limb is collectively referred to as the left hind limb and the right hind limb. For the convenience of collecting various data, various sensors are further arranged inside or on the surface of the torso 1, including but not limited to attitude sensors, limb end force sensors, laser radars or depth cameras, etc. Among them, the attitude sensor is a combination of a gyroscope and an accelerometer, which is used to detect the pitch angle, roll angle and yaw angle of the torso 1; the limb end force sensor is arranged on the sole of each front limb and hind limb, which is used to detect the contact pressure between the limb and the ground; the laser radar or depth camera is used to detect whether there is a protruding or concave terrain in front.

[0029] The running process of the above four-legged robot is shown in Figure 2 , which at least includes the following three steps. Among them, step S1: according to the preset running demand of the four-legged robot, the time proportion of the support phase and the swing phase of the left front limb, the right front limb, the left hind limb and the right hind limb in a gait cycle is determined, as well as the gait phase difference between any two limbs.

[0030] This step is essentially a gait planning process, and its essence is to optimize the allocation of the time proportion of the "support phase" and the "swing phase" and the limb phase difference according to the preset running demand of the four-legged robot, such as low-speed patrol and medium-speed operation, to balance the two core indicators of stability and efficiency. The stability priority scenario refers to the need to extend the single limb support phase time to ensure that the torso 1 has enough load bearing time, and to stagger the limb support timing to avoid imbalance caused by multiple limbs swinging at the same time, so at this time the parameters need to be set as "high support phase proportion, small phase difference". The efficiency priority scenario refers to the need to shorten the single limb support time and improve the swing frequency, and at the same time reduce the interference between the limbs through "group synchronous motion", so at this time the parameters need to be set as "balanced support phase / swing phase proportion, large group phase difference".

[0031] The support phase is the stage when the limb is in contact with the ground and bears the load of the torso, and the swing phase is the stage when the limb swings away from the ground and towards the target landing position.

[0032] The preset operation requirement includes two states of medium-speed high-efficiency operation and low-speed stable operation. When the medium-speed high-efficiency operation is adopted, diagonal gait planning is adopted, the left front limb and the right rear limb are divided into a first limb group, and the right front limb and the left rear limb are divided into a second limb group. The gait phase difference between the first limb group and the second limb group is 180°, and the support phase time ratio and the swing phase time ratio of a single limb are both 50%.

[0033] When the low-speed stable operation is adopted, walking gait planning is adopted, and the left front limb, the right rear limb, the right front limb and the left rear limb enter the support phase in sequence. The gait phase difference between the adjacent two limbs is 90°, and the support phase time ratio of a single limb is 60% and the swing phase time ratio is 40%.

[0034] S2, when the four-legged robot is in the support phase, the elbow primary drive motor 21 and the knee primary drive motor 31 output corresponding torques to support the torso 1, and at the same time, the elbow secondary drive motor 22 and the knee secondary drive motor 32 finely adjust the joint angle to adapt to the ground state; when the four-legged robot is in the swing phase, the elbow primary drive motor 21 and the knee primary drive motor 31 drive the joint flexion to drive the limb swing, and at the same time, the elbow secondary drive motor 22 and the knee secondary drive motor 32 adjust the joint swing angle to plan the predetermined swing trajectory of the limb end.

[0035] This step is essentially a division of labor and cooperation of the elbow primary drive motor 21 and the knee primary drive motor 31 responsible for basic power output, and the elbow secondary drive motor 22 and the knee secondary drive motor 32 responsible for fine attitude adjustment, and combined with the structural characteristics of the front elbow and the rear knee (the front limb supports the elbow and bears the weight, and the rear limb bends the knee and advances), the function adaptation in different movement stages is realized.

[0036] The support phase includes two stages of support phase initial stage and support phase middle stage. In the support phase initial stage, when the elbow primary drive motor 21 of the left front limb and the right front limb drives the elbow to extend, the knee primary drive motor 31 of the left rear limb and the right rear limb drives the knee to bend, and jointly receives the load of the torso 1. In the support phase middle stage, when the elbow secondary drive motor 22 of the left front limb and the right front limb adjusts the elbow swing angle, the knee secondary drive motor 32 of the left rear limb and the right rear limb synchronously adjusts the knee swing angle, and maintains the horizontal posture of the torso.

[0037] Similarly, the swing phase also includes two stages of swing phase initial stage and swing phase middle stage. In the swing phase initial stage, when the elbow primary drive motor 21 of the left front limb and the right front limb drives the elbow to bend, the knee primary drive motor 31 of the left rear limb and the right rear limb drives the knee to extend, so that the limb is separated from the ground. In the swing phase middle stage, when the elbow secondary drive motor 22 of the left front limb and the right front limb controls the limb end height from the ground in the interval of 10-15 cm, the knee secondary drive motor 32 of the left rear limb and the right rear limb controls the limb end height from the ground in the interval of 8-12 cm, so as to ensure that the limb end swings to the landing position according to the predetermined trajectory.

[0038] In order to ensure the operation effect, when the diagonal gait is adopted, the adjustment frequency of the elbow secondary drive motor 22 and the knee secondary drive motor 32 in the middle stage of the support phase is increased to 50 Hz, the single adjustment amplitude of the front limb is ≤2° or the single adjustment amplitude of the rear limb is ≤1.5°, so as to ensure that the contact pressure distribution of the limb end and the ground is uniform, and the running speed of the quadruped robot reaches 0.8-1.0 m / s±0.1 m / s.

[0039] When the walking gait is adopted, the swing accuracy of the elbow secondary drive motor 22 and the knee secondary drive motor 32 in the middle stage of the swing phase is ≤±2 cm, so as to ensure that the landing interval of the front limb and the interval of the current support point is 0.18-0.22 m, and the running speed of the quadruped robot reaches 0.25-0.35 m / s±0.05 m / s.

[0040] S3, real-time acquisition of the running state data of the quadruped robot, and dynamic adjustment of the control parameters of the elbow primary drive motor 21, the elbow secondary drive motor 22, and the knee primary drive motor 31 and the knee secondary drive motor 32 according to the running state data, so as to ensure that the posture of the trunk 1 of the quadruped robot is in a stable state.

[0041] The control parameters in this step include the following modes: If the pitch angle fluctuation of the trunk 1 exceeds 1°, the output torque of the elbow primary drive motor 21 of the left front limb and the right front limb in the support phase is increased or the extension speed of the knee primary drive motor 31 of the left rear limb and the right rear limb in the swing phase is adjusted.

[0042] If the contact pressure of any limb of the left front limb, the right front limb, the left rear limb and the right rear limb exceeds the preset threshold, the joint swing angle of the corresponding limb is fine-tuned through the elbow secondary drive motor 22 or the knee secondary drive motor 32, so as to disperse the load of the trunk 1.

[0043] The preset threshold mainly includes a trunk posture threshold and a limb end contact pressure threshold. The trunk posture data includes a pitch angle (forward and backward inclination), a roll angle (left and right inclination), and a yaw angle (turning deviation). These data directly reflect the overall stability of the quadruped robot, and the threshold setting needs to take "not breaking the support triangle structure of the front elbow and the rear knee" as the core target. The limb end contact pressure is directly related to the motor load and the ground adaptability, and the threshold setting needs to distinguish between the front limbs and the rear limbs (due to the difference in the load bearing function of the front elbow and the rear knee structure) and combine with the rated load capacity of the motor.

[0044] In the embodiment, the static stability threshold of the pitch angle is ±1°, and the dynamic change rate threshold is ±0.5° / s; the static stability threshold of the roll angle is ±0.5°, and the dynamic change rate threshold is ±0.3° / s; the turning process threshold of the yaw angle is ±2°, and the straight running process threshold is ±0.5°.

[0045] The static support threshold of the front limb is 150-200 N, and the peak protection threshold is ≤250 N; the static support threshold of the rear limb is 200-250 N, and the peak protection threshold is ≤300 N.

[0046] It should be further explained that in addition to the above steps S1-S3, the embodiment further includes a terrain adaptability adjustment step S4 and a turning control step S5.

[0047] The terrain adaptability adjustment is a core adaptive link for the stable operation of the quadruped robot in uneven environments such as convex, concave, and slope scenes. It is based on the structural characteristics of the front elbow and the rear knee and the division advantages of the two-stage motor, and solves the problems of trunk jolting and support failure caused by terrain height difference through "advance detection-accurate parameter adjustment-real-time correction" closed-loop control, which is a scenario-based supplement to steps S2 and S3. Specifically, when the quadruped robot detects a convex or concave terrain in front through the laser radar or depth camera carried by the trunk 1, the secondary driving motor of the corresponding limb is controlled to adjust the joint swing angle 100-150 ms before the limb enters the support phase. That is, if the left front limb or the right front limb contacts a convex terrain, the elbow secondary driving motor 22 of the left front limb or the right front limb swings to the convex side, and the extension speed of the elbow primary driving motor 21 of the left front limb or the right front limb is reduced at the same time. At this time, the swing angle of the elbow secondary driving motor 22 of the front limb is controlled to be 8-10°, and the speed of the elbow primary driving motor 21 of the front limb is reduced by 20%-40%.

[0048] If the left hind leg or the right hind leg contacts the concave terrain, the knee secondary drive motor 32 of the left hind leg or the right hind leg turns to the concave side, and synchronously reduces the bending speed of the knee primary drive motor 31 of the left hind leg or the right hind leg. At this time, the turning angle of the knee secondary drive motor 32 of the hind leg to the concave side is 6-8°, and the bending speed of the knee primary drive motor 31 of the hind leg is reduced by 15%-30%, ensuring that the body pitch amplitude is ≤3cm.

[0049] The turning control is a link for the quadruped robot to realize flexible path adjustment in a complex scene. Based on the structural characteristics of the "front elbow and rear knee" and the division advantages of the two-stage motors, the turning control forms motion differential through "turning side trajectory optimization + non-turning side power enhancement", realizes smooth and accurate direction adjustment without additional turning mechanism.

[0050] Specifically, S5, when the quadruped robot needs to turn left, the elbow secondary drive motor 22 of the left front leg and the knee secondary drive motor 32 of the left hind leg increase the inside swing range, and the elbow primary drive motor 21 of the right front leg and the knee primary drive motor 31 of the right hind leg increase the driving speed to form motion differential.

[0051] When the quadruped robot needs to turn right, the elbow secondary drive motor 22 of the right front leg and the knee secondary drive motor 32 of the right hind leg increase the inside swing range, and the elbow primary drive motor 21 of the left front leg and the knee primary drive motor 31 of the left hind leg increase the driving speed to form motion differential.

[0052] To ensure the running effect, in the embodiment, for the front leg and the hind leg on the turning side, the elbow secondary drive motor 22 of the front leg needs to increase the inside swing range by 12%-15%, and the knee secondary drive motor 32 of the hind leg needs to increase the inside swing range by 10%-12%; for the front leg and the hind leg on the non-turning side, the elbow primary drive motor 21 of the front leg and the knee primary drive motor 31 of the hind leg need to increase the driving speed by 8%-12%, forming the motion differential between the turning side and the non-turning side, so that the turning radius of the quadruped robot is controlled within 0.5-0.6m±0.1m, and the body 1 yaw angle fluctuation is within ≤2° during the turning process.

[0053] It should be noted that the above steps S4 and S5 do not have a fixed sequence, and the quadruped robot in the embodiment can call step S4 or S5 at any time after executing step S3 to meet its immediacy requirement.

[0054] As described above, the application can be well implemented.

Claims

1. A method for operating a quadruped robot driven by two-stage motors (elbow and knee), the quadruped robot comprising a torso, a left and right forelimbs (both elbow-type structures) located at the front of the torso, and a left and right hindlimbs (both knee-type structures) located at the rear of the torso, wherein the left and right forelimbs each have an elbow-type primary drive motor and an elbow-type secondary drive motor internally, and the left and right hindlimbs each have a knee-type primary drive motor and a knee-type secondary drive motor internally, characterized in that... Its operation method includes the following steps: S1. Based on the preset operating requirements of the quadruped robot, determine the time proportions of the support phase and swing phase of each of the left forelimb, right forelimb, left hindlimb, and right hindlimb in a gait cycle, as well as the gait phase difference between any two limbs; S2. When the quadruped robot is in the support phase, the elbow primary drive motor and the knee primary drive motor output stress torque to support the torso. At the same time, the elbow secondary drive motor and the knee secondary drive motor fine-tune the joint angle to adapt to the ground condition. When the quadruped robot is in the swing phase, the elbow primary drive motor and the knee primary drive motor drive the joint to flex and extend to drive the limb to swing. At the same time, the elbow secondary drive motor and the knee secondary drive motor adjust the joint rotation angle to plan the predetermined swing trajectory of the limb. S3. Acquire the running status data of the quadruped robot in real time, and dynamically adjust the control parameters of each elbow primary drive motor, elbow secondary drive motor, knee primary drive motor, and knee secondary drive motor based on the running status data to ensure that the quadruped robot's torso posture is in a stable state.

2. The method for operating a quadruped robot driven by a two-stage motor (front elbow, rear knee) according to claim 1, characterized in that, When the preset running requirement mentioned in step S1 is to run at medium speed and high efficiency, it adopts diagonal gait planning, dividing the left forelimb and right hindlimb into the first limb group, and the right forelimb and left hindlimb into the second limb group; the gait phase difference between the first limb group and the second limb group is 180°, and the proportion of the support phase time and the swing phase time of a single limb are both 50%.

3. The method for operating a quadruped robot driven by a two-stage motor (front elbow, rear knee) according to claim 1, characterized in that, In step S1, the preset running requirement is that when running at low speed and stability, walking gait planning is adopted. The left forelimb, right hindlimb, right forelimb, and left hindlimb enter the support phase in sequence. The gait phase difference between adjacent limbs is 90°, and the support phase time of a single limb accounts for 60% and the swing phase time accounts for 40%.

4. A method for operating a quadruped robot driven by a two-stage motor (front elbow, rear knee) according to any one of claims 1 to 3, characterized in that, When the quadruped robot is in the support phase in step S2, it includes the initial support phase and the middle support phase. In the initial support phase, when the elbows of the left and right forelimbs are driven by the first-stage drive motors to extend the elbows, the knees of the left and right hindlimbs are driven by the first-stage drive motors to bend the knees, thus jointly bearing the load of the torso. During the middle of the support phase, when the elbow secondary drive motors of the left and right forelimbs adjust the elbow swing angle, the knee secondary drive motors of the left and right hindlimbs simultaneously adjust the knee swing angle to maintain the horizontal posture of the trunk.

5. The method for operating a quadruped robot driven by a two-stage motor (front elbow, rear knee) according to claim 4, characterized in that, When the quadruped robot is in the swing phase in step S2, it includes the initial swing phase and the middle swing phase. In the initial swing phase, when the elbow primary drive motors of the left and right forelimbs drive the elbows to bend, the knee primary drive motors of the left and right hindlimbs drive the knees to extend, causing the limbs to lift off the ground. In the middle swing phase, when the elbow secondary drive motors of the left and right forelimbs control the height of the limbs off the ground in the range of 10 to 15 cm, the knee secondary drive motors of the left and right hindlimbs control the height of the limbs off the ground in the range of 8 to 12 cm, ensuring that the limbs swing towards the landing position according to the predetermined trajectory.

6. The method for operating a quadruped robot driven by a two-stage motor (front elbow, rear knee) according to claim 5, characterized in that, The specific method for "dynamically adjusting the control parameters of each elbow primary drive motor, elbow secondary drive motor, knee primary drive motor, and knee secondary drive motor according to the operating status data" in step S3 is as follows: if the torso pitch angle fluctuates more than 1°, the output torque of the elbow primary drive motor in the support phase of the left and right forelimbs is increased or the extension speed of the knee primary drive motor in the swing phase of the left and right hindlimbs is adjusted; if the contact pressure of any limb of the left forelimb, right forelimb, left hindlimb, and right hindlimb exceeds a preset threshold, the joint rotation angle is finely adjusted by the secondary drive motor of the corresponding limb to distribute the torso load.

7. The method for operating a quadruped robot driven by a two-stage motor (front elbow, rear knee) according to claim 4, characterized in that, The operating method also includes terrain-adaptive adjustments. When the quadruped robot detects a raised or recessed terrain in front of it, if the left or right forelimb is in contact with the raised terrain, the elbow secondary drive motor of the left or right forelimb swings towards the raised side, and the extension speed of the elbow primary drive motor of the left or right forelimb is reduced simultaneously. If the left or right hindlimb is in contact with the recessed terrain, the knee secondary drive motor of the left or right hindlimb swings towards the recessed side, and the bending speed of the knee primary drive motor of the left or right hindlimb is reduced simultaneously.

8. The method for operating a quadruped robot driven by a two-stage motor (front elbow, rear knee) according to claim 4, characterized in that, The operation method also includes steering control. When the quadruped robot turns left, the elbow secondary drive motor of the left forelimb and the knee secondary drive motor of the left hindlimb increase the inward swing amplitude, and the elbow primary drive motor of the right forelimb and the knee primary drive motor of the right hindlimb both increase the drive speed to form a motion differential. When the quadruped robot turns right, the elbow secondary drive motor of the right forelimb and the knee secondary drive motor of the right hindlimb increase the inward swing amplitude, and the elbow primary drive motor of the left forelimb and the knee primary drive motor of the left hindlimb both increase the drive speed to form a motion differential.