Tendon-driven intelligent humanoid robot with body and control system

Abstract

The invention discloses a tendon-driven intelligent humanoid robot with a body and a control system, and belongs to the technical field of robots. The problems that an existing humanoid robot is often stiff during joint movement, and the flexible movement of human muscles is difficult to simulate are solved, and by dynamically adjusting the frequency, the amplitude and the phase of a driving signal and combining a self-adaptive frequency algorithm and a closed-loop control strategy, the flexible movement of the human muscles is simulated. It is ensured that the humanoid robot body can achieve quick response and accurate motion control in different task scenes; a phase synchronization algorithm is adopted to dynamically adjust the phase of a multi-joint driving signal, the execution efficiency of a multi-joint cooperative task is optimized, the fluency and stability of the motion of the humanoid robot body are enhanced, and the humanoid robot body can complete complex motions more efficiently; and the weight coefficient of the control parameter is dynamically adjusted according to the real-time motion state and the environment information, so that the humanoid robot body can flexibly adapt to different task scenes and load conditions.

Description

Technical Field

[0001] This invention relates to the field of embodied intelligent robot technology, specifically to a tendon-driven embodied intelligent humanoid robot and its control system. Background Technology

[0002] With the rapid development of artificial intelligence and robotics, humanoid robots are showing broad application prospects in many fields such as industry, service, and healthcare.

[0003] Currently, the driving methods for humanoid robots mainly include motor drive and hydraulic drive. Although these driving methods can achieve certain motion functions, they still have shortcomings in terms of flexibility, naturalness and safety. For example, motor-driven humanoid robots are often stiff when moving their joints and it is difficult to simulate the smooth movement of human muscles.

[0004] Therefore, to meet current needs, a tendon-driven embodied intelligent humanoid robot and its control system are proposed. Summary of the Invention

[0005] The purpose of this invention is to provide a tendon-driven intelligent humanoid robot and its control system. By dynamically adjusting the frequency, amplitude, and phase of the drive signal, combined with an adaptive frequency algorithm and a closed-loop control strategy, the humanoid robot body can achieve rapid response and precise motion control in different task scenarios. A phase synchronization algorithm is used to dynamically adjust the phase of the drive signals of multiple joints, optimizing the execution efficiency of multi-joint collaborative tasks, enhancing the smoothness and stability of the humanoid robot body's movements, and enabling it to complete complex actions more efficiently. Furthermore, the weighting coefficients of the control parameters are dynamically adjusted according to real-time motion status and environmental information, allowing the humanoid robot body to flexibly adapt to different task scenarios and load conditions, thus solving the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] A tendon-driven intelligent humanoid robot includes: a humanoid robot body, which is composed of a head, upper body, shoulder joint, elbow joint, wrist joint, waist joint, thigh, knee joint, calf, ankle joint and foot. The wrist joint is connected to the palm. Multiple tendon units are provided inside the shoulder joint, elbow joint, wrist joint, waist joint, knee joint, ankle joint and palm. Each tendon unit is composed of an artificial tendon, an actuator and a tendon tension sensor, and is configured to simulate the mechanical properties of a real tendon through the artificial tendon.

[0008] The artificial tendon is controlled by a actuator to contract and relax, driving the joints of the hand to move in multiple directions.

[0009] The tension and elongation parameters of the artificial tendon are monitored in real time using a tendon tension sensor.

[0010] Furthermore, the front of the head is equipped with a depth camera for real-time monitoring of the environment around the humanoid robot, and a lidar for acquiring precise distance and angle information of the surrounding environment and forming point cloud data.

[0011] Furthermore, the head is connected to the upper body, and the two sides of the upper body are connected to the left and right elbow joints respectively through two shoulder joints. Each elbow joint is connected to the left and right arms respectively through the corresponding wrist joint. The lower part of the upper body is connected to the waist joint, and the lower part of the waist joint is connected to the two thighs respectively. Each thigh is connected to the lower leg through the corresponding knee joint, and the lower part of the lower leg is connected to the foot through the corresponding ankle joint.

[0012] A tendon-driven android control system includes:

[0013] The data collection unit is configured to acquire real-time state information of the humanoid robot body through various sensors installed on various parts of the humanoid robot body, including but not limited to force sensors and tactile sensors, as well as depth cameras and lidar.

[0014] The intelligent control unit is configured to combine pre-set task objectives and behavior rules, use a neural network model to make autonomous decisions, and generate corresponding motion commands and behavior strategies.

[0015] The generated instructions are converted into tendon-driven signals to drive the movement and behavior of the humanoid robot body.

[0016] Furthermore, the intelligent control unit includes:

[0017] The data fusion module is configured to introduce multimodal data fusion technology, which integrates object position information collected by visual sensors, contact force information collected by force sensors, and contact information collected by tactile sensors to accurately calculate the degree of tendon contraction and relaxation.

[0018] The instruction conversion module is configured to adjust the frequency, amplitude, and phase of the generated instructions based on the motion state of the humanoid robot and the task requirements.

[0019] Furthermore, the instruction conversion module adjusts the frequency, amplitude, and phase of the generated instructions based on the motion state of the humanoid robot and the task requirements, including:

[0020] Based on the priority and complexity of the task, the module presets different frequency ranges; for high-priority tasks, a high-frequency driving strategy is used first to ensure a fast response.

[0021] It also employs an adaptive frequency algorithm to dynamically adjust the frequency of the drive signal based on real-time motion data fed back from the sensor;

[0022] Based on the preset motion trajectory and task objectives, the amplitude of the drive signal is dynamically adjusted to optimize the motion performance of the hand joints;

[0023] It also adopts a closed-loop control strategy to adjust the amplitude of the drive signal in real time based on the real-time feedback load data;

[0024] In multi-joint collaborative tasks, a phase synchronization algorithm is adopted to dynamically adjust the phase of the driving signal by analyzing the motion trajectory and time series data of each joint of the palm, so as to ensure the coordinated movement between multiple tendon units.

[0025] Based on the data collected by the vision sensor, the phase of the drive signal is adjusted in real time to adapt to the dynamic environment.

[0026] Furthermore, the intelligent control unit also includes:

[0027] The weight adjustment module is configured to analyze the tension and elongation data collected by the tendon unit, and dynamically adjust the weight coefficients of the control parameters in combination with the robot's motion state and environmental information.

[0028] When the robot is under high load or in a complex motion state, the weighting coefficient of the tension data is automatically increased; when the humanoid robot body is under low load or in a simple motion state, the weighting coefficient of the tension data is decreased.

[0029] It also automatically adjusts the weighting coefficients based on different task scenarios and environmental conditions.

[0030] Furthermore, the intelligent control unit also includes:

[0031] The energy management module is configured to monitor the energy consumption of the humanoid robot in real time and dynamically adjust the energy allocation according to task priority and motion status.

[0032] The self-healing module is configured to monitor the operating status of each component of the humanoid robot in real time and automatically start the diagnostic program when a fault is detected.

[0033] Furthermore, it also includes:

[0034] The environmental interaction unit is configured to acquire and respond to human voice commands through speech recognition and speech synthesis technologies.

[0035] By recognizing human hand gestures using depth cameras and image processing algorithms, and combining this with the visual perception and force feedback of the humanoid robot itself, the robot can precisely grasp, move, and manipulate objects.

[0036] Furthermore, it also includes:

[0037] The human-robot collaboration unit is configured to provide a visual interface, support collaborative tasks between the humanoid robot and the human operator, and realize task allocation and collaborative control through various interaction methods.

[0038] The safety monitoring unit is configured to monitor the operating status of the humanoid robot and its surrounding environment in real time through the data collection unit, ensuring that it operates within a safe range.

[0039] Compared with the prior art, the beneficial effects of the present invention are:

[0040] In this invention, by dynamically adjusting the frequency, amplitude, and phase of the driving signals, combined with an adaptive frequency algorithm and a closed-loop control strategy, the humanoid robot body can achieve rapid response and precise motion control in different task scenarios, significantly improving the smoothness of joint movement and the accuracy of task execution. A phase synchronization algorithm is employed to dynamically adjust the phase of the driving signals for multiple joints, optimizing the execution efficiency of multi-joint collaborative tasks and enhancing the smoothness and stability of the humanoid robot body's movements, enabling it to complete complex actions more efficiently. Furthermore, the weighting coefficients of the control parameters are dynamically adjusted based on real-time motion status and environmental information, allowing the humanoid robot body to flexibly adapt to different task scenarios and load conditions, improving the robustness of the control system and the success rate of task execution. Attached Figure Description

[0041] Figure 1 This is a structural diagram of the tendon-driven embodied intelligent humanoid robot of the present invention;

[0042] Figure 2 This is a diagram showing the distribution of tendon drives in the palm of the hand according to the present invention.

[0043] Figure 3 This is a diagram of the tendon-driven embodied intelligent humanoid robot control system of the present invention.

[0044] In the image: 1. Head; 2. Upper body; 3. Shoulder joint; 4. Elbow joint; 5. Wrist joint; 6. Waist joint; 7. Thigh; 8. Knee joint; 9. Lower leg; 10. Ankle joint; 11. Foot; 12. Palm; 13. Artificial tendon; 14. Actuator; 15. Tendon tension sensor; 16. Depth camera; 17. LiDAR. Detailed Implementation

[0045] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0046] To address the shortcomings of existing humanoid robots in terms of flexibility, naturalness, and safety—such as the often stiff joint movements of motor-driven humanoid robots, which make it difficult to simulate the smooth movements of human muscles—please refer to [the relevant documentation / reference]. Figures 1-3 This embodiment provides the following technical solution:

[0047] A tendon-driven embodied intelligent humanoid robot includes: a humanoid robot body, which is composed of a head 1, an upper body 2, shoulder joints 3, elbow joints 4, wrist joints 5, waist joints 6, thighs 7, knee joints 8, lower legs 9, ankle joints 10, and feet 11. The head 1 is connected to the upper body 2. The two sides of the upper body 2 are connected to the left and right elbow joints 4 via two shoulder joints 3 respectively. Each elbow joint 4 is connected to the left and right arms via corresponding wrist joints 5 respectively. The lower part of the upper body 2 is connected to the waist joint 6. The lower part of the waist joint 6 is connected to two thighs 7 respectively. Each thigh 7 is connected to the lower leg 9 via a corresponding knee joint 8. The lower part of the lower leg 9 is connected to the foot 11 via a corresponding ankle joint 10. The wrist joints 5 are connected to the palms 12. The shoulder joints 3, elbow joints 4, wrist joints 5, waist joints 6, knee joints 8, ankle joints 10, and palms are also connected. Each of the 12 has a tendon unit inside; multiple tendon units are composed of artificial tendons 13, actuators 14 and tendon tension sensors 15. The artificial tendons 13 are made of materials with high elasticity, high strength and low hysteresis characteristics, such as shape memory alloy fibers, electroactive polymers, etc., and the mechanical properties of real tendons are simulated by the artificial tendons 13. The contraction and relaxation of the artificial tendons 13 are controlled by the actuators 14 to drive the joints of the hand 12 to move in multiple directions. The tension and elongation parameters of the artificial tendons 13 are monitored in real time by the tendon tension sensors 15 to provide data support for the feedback adjustment of the control system. The front of the head 1 is equipped with a depth camera 16 for real-time monitoring of the environment around the humanoid robot body, and a lidar 17 for acquiring precise distance and angle information of the surrounding environment and forming point cloud data.

[0048] A tendon-driven android control system includes:

[0049] The data collection unit is configured to acquire real-time state information of the humanoid robot body through various sensors installed on various parts of the humanoid robot body, including but not limited to force sensors and tactile sensors, as well as depth camera 16 and lidar 17.

[0050] The intelligent control unit is configured to combine pre-set task objectives and behavioral rules, use a neural network model to make autonomous decisions, generate corresponding motion commands and behavioral strategies, and convert the generated commands into tendon drive signals to drive and control the movement and behavior of the humanoid robot body.

[0051] The intelligent control unit includes:

[0052] The data fusion module is configured to incorporate multimodal data fusion technology, which integrates object position information collected by visual sensors, contact force information collected by force sensors, and contact information collected by tactile sensors to accurately calculate the degree of tendon contraction and relaxation, thereby achieving precise grasping and manipulation of objects.

[0053] The instruction conversion module is configured to adjust the frequency, amplitude, and phase of the generated instructions based on the motion state and task requirements of the humanoid robot, ensuring the accuracy and real-time performance of tendon actuation, thereby achieving compliant joint movement and efficient task execution of the humanoid robot. Specifically, it includes:

[0054] Based on the priority and complexity of the tasks, the module presets different frequency ranges. For high-priority tasks, a high-frequency drive strategy is used to ensure rapid response. For example, when performing slow and stable motion tasks, such as fine manipulation or load handling, the frequency of the drive signal is adjusted to a lower level to ensure stable contraction and relaxation of tendons and avoid excessive vibration and energy waste. When performing tasks requiring rapid response, such as obstacle avoidance or rapid grasping, the frequency of the drive signal is increased to achieve rapid tendon contraction and joint movement, ensuring the agility of the humanoid robot. An adaptive frequency algorithm is also used to dynamically adjust the frequency of the drive signal based on real-time motion data from sensors, such as joint velocity and acceleration. For example, when an increase in joint movement velocity is detected, the adaptive frequency algorithm automatically increases the frequency of the drive signal to match the motion requirements.

[0055] Based on the preset motion trajectory and task objectives, the amplitude of the drive signal is dynamically adjusted to optimize the motion performance of the 12 joints of the hand. For example, in tasks requiring a wide range of motion, such as arm extension or leg stepping, the amplitude is increased to achieve a wider range of joint movements; in tasks requiring fine manipulation, such as finger grasping, the amplitude is reduced to achieve more precise control. A closed-loop control strategy is adopted to adjust the amplitude of the drive signal in real time based on the real-time feedback load data, enabling it to respond quickly to load changes and ensuring the accuracy of the drive signal amplitude.

[0056] In multi-joint collaborative tasks, a phase synchronization algorithm is employed to dynamically adjust the phase of the drive signals by analyzing the motion trajectories and time-series data of each of the 12 joints of the hand, ensuring coordinated movement among multiple tendon units. For example, in the arm grasping task, the module adjusts the phase of the drive signals for the shoulder joint 3, elbow joint 4, and wrist joint 5 to ensure coordinated movement of each joint and achieve a smooth grasping action. Based on the data collected by the force sensor, the phase of the drive signals is adjusted in real time to adapt to the dynamic environment. For example, during walking, the phase of the drive signals for the knee joint 8 and ankle joint 10 is dynamically adjusted according to ground conditions and obstacle positions to ensure walking stability and safety.

[0057] The weight adjustment module is configured to analyze the tension and elongation data collected by the tendon unit, and dynamically adjust the weight coefficients of the control parameters in combination with the robot's motion state and environmental information. When the robot is under high load or complex motion, the weight coefficient of the tension data is automatically increased to ensure the accuracy and stability of the tendon drive signal. When the humanoid robot body is under low load or simple motion, the weight coefficient of the tension data is decreased to optimize energy consumption and improve response speed. It also automatically adjusts the weight coefficient adjustment strategy according to different task scenarios and environmental conditions to improve the adaptability and operational accuracy of the humanoid robot body in complex environments.

[0058] The beneficial effects achieved by the above are as follows: By dynamically adjusting the frequency, amplitude, and phase of the driving signals, combined with adaptive frequency algorithms and closed-loop control strategies, the humanoid robot body can achieve rapid response and precise motion control in different task scenarios, significantly improving the smoothness of joint movement and the accuracy of task execution; by adopting a phase synchronization algorithm to dynamically adjust the phase of the driving signals of multiple joints, the execution efficiency of multi-joint collaborative tasks is optimized, enhancing the smoothness and stability of the humanoid robot body's movements, enabling it to complete complex actions more efficiently; and by dynamically adjusting the weight coefficients of the control parameters according to real-time motion status and environmental information, the humanoid robot body can flexibly adapt to different task scenarios and load conditions, improving the robustness of the control system and the success rate of task execution.

[0059] The energy management module is configured to monitor the energy consumption of the humanoid robot in real time and dynamically adjust energy allocation based on task priority and motion status. For example, in high-load or complex task scenarios, energy is prioritized for the tendon units of key joints to ensure the smooth completion of the task. In low-load or standby states, energy supply is reduced to optimize energy consumption and extend the battery life of the humanoid robot. Combined with the task planning of the intelligent control unit, energy demand is predicted and energy allocation strategies are adjusted in advance to ensure the stable operation of the humanoid robot in continuous tasks.

[0060] The self-healing module is configured to monitor the operating status of each component of the humanoid robot in real time and automatically start the diagnostic program when a fault is detected. For example, for sensor faults, it uses data from other sensors for compensation and adjusts the control signal through an adaptive algorithm to ensure the safety and reliability of the humanoid robot in fault conditions.

[0061] The environmental interaction unit is configured to acquire and respond to human voice commands through speech recognition and speech synthesis technologies, enabling natural language communication between the humanoid robot and humans; it recognizes human gestures through a depth camera 16 and image processing algorithms to achieve non-contact human-computer interaction; and combined with the visual perception and force feedback of the humanoid robot, it controls the precise grasping, handling, and manipulation of objects, improving the operational capabilities and adaptability of the humanoid robot in practical applications.

[0062] The human-robot collaboration unit is configured to provide a visual interface to support collaborative tasks between the humanoid robot and the human operator. It enables task allocation and collaborative control through various interaction methods. For example, the visual interface allows the operator to monitor the status of the humanoid robot in real time and manually adjust task parameters. Combined with voice interaction, it enables real-time voice command interaction between the operator and the humanoid robot, improving collaboration efficiency. During collaborative tasks, the unit dynamically adjusts the humanoid robot's motion strategy based on the operator's actions and intentions to ensure the smooth progress of the task and the safety of the operator.

[0063] The safety monitoring unit is configured to monitor the operating status of the humanoid robot and its surrounding environment in real time through the data collection unit, ensuring that it operates within a safe range. For example, it uses force and tactile sensors to detect the contact force between the humanoid robot and humans or other objects in real time, preventing accidental collisions and injuries. It uses sensors such as depth camera 16 and lidar 17 to monitor the dynamic changes of the surrounding environment in real time, provide early warnings of potential dangers, and adjust its movement strategy. When a safety risk is detected, it automatically initiates an emergency braking procedure or switches to a safety mode to ensure the safety of the humanoid robot and surrounding personnel.

[0064] The beneficial effects achieved by the above are as follows: By monitoring the energy consumption of the humanoid robot body in real time and dynamically adjusting the energy allocation according to task priority and motion status, the energy utilization efficiency is optimized, the endurance of the humanoid robot body is extended, and its applicability in long-term or continuous tasks is enhanced; at the same time, the operating status of each component of the humanoid robot body is monitored in real time, and the diagnostic program is automatically started when a fault is detected, ensuring the basic functions and safety of the robot in fault conditions, and significantly reducing maintenance costs and downtime.

[0065] Working principle: The tendon unit, composed of artificial tendon 13, actuator 14, and tendon tension sensor 15, simulates the working mode of human muscles and tendons. Multiple sensors installed on various parts of the humanoid robot body collect its own status and environmental information in real time. Combining preset task objectives and behavior rules, the robot uses a neural network model to make autonomous decisions and generate motion commands. It also dynamically adjusts the frequency, amplitude, and phase of the drive signals to achieve compliant joint movement and efficient task execution. At the same time, the weight coefficients of the control parameters are dynamically optimized based on real-time data to adapt to different task scenarios and environmental conditions, enabling the humanoid robot body to complete diverse tasks efficiently and safely in complex environments.

[0066] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or high-voltage switchgear that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or high-voltage switchgear.

[0067] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A tendon-driven, embodied intelligent humanoid robot, comprising: The humanoid robot body is characterized in that the humanoid robot body is composed of a head (1), upper body (2), shoulder joint (3), elbow joint (4), wrist joint (5), waist joint (6), thigh (7), knee joint (8), calf (9), ankle joint (10) and foot (11). The wrist joint (5) is connected to the palm (12). The shoulder joint (3), elbow joint (4), wrist joint (5), waist joint (6), knee joint (8), ankle joint (10) and palm (12) are all provided with multiple tendon units. The tendon unit is composed of an artificial tendon (13) actuator (14) and a tendon tension sensor (15), and is configured to simulate the mechanical properties of a real tendon through the artificial tendon (13). The artificial tendon (13) is controlled by the actuator (14) to contract and relax, thereby driving the joints of the palm (12) to move in multiple directions; The tension and elongation parameters of the artificial tendon (13) are monitored in real time by a tendon tension sensor (15).

2. The tendon-driven, body-worn intelligent humanoid robot according to claim 1, characterized in that, The front of the head (1) is equipped with a depth camera (16) for real-time monitoring of the environment around the humanoid robot body, and a lidar (17) for acquiring precise distance and angle information of the surrounding environment and forming point cloud data.

3. The tendon-driven intelligent humanoid robot according to claim 1, characterized in that, The head (1) is connected to the upper body (2). The two sides of the upper body (2) are connected to the left and right elbow joints (4) respectively through two shoulder joints (3). Each elbow joint (4) is connected to the left and right arms respectively through the corresponding wrist joint (5). The lower part of the upper body (2) is connected to the waist joint (6). The lower part of the waist joint (6) is connected to the two thighs (7). Each thigh (7) is connected to the lower leg (9) through the corresponding knee joint (8). The lower part of the lower leg (9) is connected to the foot (11) through the corresponding ankle joint (10).

4. A tendon-driven automaton intelligent humanoid robot control system, implemented in any one of claims 1-3, characterized in that, include: The data collection unit is configured to acquire the humanoid robot's own state information in real time through various sensors installed on various parts of the humanoid robot's body, including but not limited to force sensors and tactile sensors, as well as depth cameras (16) and lidar (17). The intelligent control unit is configured to combine pre-set task objectives and behavior rules, use a neural network model to make autonomous decisions, and generate corresponding motion commands and behavior strategies. The generated instructions are converted into tendon-driven signals to drive and control the movement and behavior of the humanoid robot body.

5. A tendon-driven intelligent humanoid robot control system according to claim 4, characterized in that, The intelligent control unit includes: The data fusion module is configured to introduce multimodal data fusion technology, which integrates object position information collected by visual sensors, contact force information collected by force sensors, and contact information collected by tactile sensors to accurately calculate the degree of tendon contraction and relaxation. The instruction conversion module is configured to adjust the frequency, amplitude, and phase of the generated instructions based on the motion state of the humanoid robot and the task requirements.

6. A tendon-driven intelligent humanoid robot control system according to claim 5, characterized in that, The instruction conversion module adjusts the frequency, amplitude, and phase of the generated instructions according to the robot's motion state and task requirements, including: Based on the priority and complexity of the task, the module presets different frequency ranges; for high-priority tasks, a high-frequency driving strategy is used first to ensure a fast response. It also employs an adaptive frequency algorithm to dynamically adjust the frequency of the drive signal based on real-time motion data fed back from the sensor; Based on the preset motion trajectory and task objectives, the amplitude of the driving signal is dynamically adjusted to optimize the motion performance of the palm (12) joint; It also adopts a closed-loop control strategy to adjust the amplitude of the drive signal in real time based on the real-time feedback load data; In multi-joint collaborative tasks, a phase synchronization algorithm is adopted to dynamically adjust the phase of the driving signal by analyzing the motion trajectory and time series data of each joint of the palm (12) to ensure the coordinated movement between multiple tendon units; Based on the data collected by the vision sensor, the phase of the drive signal is adjusted in real time to adapt to the dynamic environment.

7. A tendon-driven intelligent humanoid robot control system according to claim 5, characterized in that, The intelligent control unit also includes: The weight adjustment module is configured to analyze the tension and elongation data collected by the tendon unit, and dynamically adjust the weight coefficients of the control parameters in combination with the robot's motion state and environmental information. When the robot is under high load or in a complex motion state, the weighting coefficient of the tension data is automatically increased; when the humanoid robot body is under low load or in a simple motion state, the weighting coefficient of the tension data is decreased. The system automatically adjusts the weighting coefficients based on different task scenarios and environmental conditions.

8. A tendon-driven intelligent humanoid robot control system according to claim 7, characterized in that, The intelligent control unit also includes: The energy management module is configured to monitor the energy consumption of the humanoid robot in real time and dynamically adjust the energy allocation according to task priority and motion status. The self-healing module is configured to monitor the operating status of each component of the humanoid robot in real time and automatically start the diagnostic program when a fault is detected.

9. A tendon-driven intelligent humanoid robot control system according to claim 4, characterized in that, Also includes: The environmental interaction unit is configured to acquire and respond to human voice commands through speech recognition and speech synthesis technologies. By recognizing human hand gestures using a depth camera (16) and image processing algorithms, and combining the visual perception and force feedback of the humanoid robot body, the robot can control the precise grasping, handling, and manipulation of objects.

10. A tendon-driven intelligent humanoid robot control system according to claim 9, characterized in that, Also includes: The human-robot collaboration unit is configured to provide a visual interface, support collaborative tasks between the humanoid robot and the human operator, and realize task allocation and collaborative control through various interaction methods. The safety monitoring unit is configured to monitor the operating status of the humanoid robot and its surrounding environment in real time through the data collection unit, ensuring that it operates within a safe range.

Images