Footed Robot Landing Control via Relative Speed and Joint Torque
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for reducing the impact of a footed robot's foot portion against the ground are costly and require high accuracy in kinematics planning, making them inefficient and expensive to implement.
Innovation Solution
A computer-implemented method for footed robot landing control that detects the landing motion state, calculates relative speeds, and controls joint torques based on contact forces to minimize the impact during the landing process, using sensors like torque sensors and laser sensors to determine the optimal motion trajectories and forces.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If inverse kinematics planning with high-accuracy planned curves is used to reduce foot impact, then the impact reduction effectiveness is improved, but the device complexity and cost increase due to higher requirements on planning accuracy and driving performance
Solution Approach 1:
The patent replaces complex mechanical inverse kinematics planning with a simplified control method based on relative speed detection and joint torque adjustment. Instead of using high-accuracy planned curves that require complex kinematics calculations, the system directly detects landing motion states and adjusts joint torques in real-time to reduce impact, substituting mechanical planning complexity with sensor-based feedback control
Solution Approach 2:
The robot system performs self-adjustment during landing by detecting its own motion state through sensors and automatically adjusting joint torques to minimize impact. The system serves itself by using onboard sensors (accelerometers, torque sensors) to monitor landing progression and autonomously modify control parameters without requiring external high-accuracy trajectory planning
2Object-affected harmful factors
If inverse kinematics planning with high-accuracy planned curves is used to reduce foot impact, then the impact reduction effectiveness is improved, but the ease of operation deteriorates due to higher requirements on planning accuracy and driving performance
Solution Approach 1:
The patent replaces complex mechanical inverse kinematics planning with a simplified control method based on relative speed detection and joint torque adjustment. Instead of using high-accuracy planned curves that require complex kinematics calculations, the system directly detects landing motion states and adjusts joint torques in real-time to reduce impact, substituting mechanical planning complexity with sensor-based feedback control
Solution Approach 2:
The system implements continuous feedback control by detecting landing motion states through sensors (accelerometers, torque sensors), comparing detected states with desired landing conditions, and adjusting joint torques accordingly. This feedback mechanism simplifies operation by automatically adapting to varying landing conditions without requiring pre-computed high-accuracy trajectories
3Ease of manufacture
If conventional impact reduction methods are avoided to reduce cost, then the manufacturing cost is reduced, but the impact reduction effectiveness deteriorates
Solution Approach 1:
The robot system performs self-adjustment during landing by detecting its own motion state through sensors and automatically adjusting joint torques to minimize impact. The system serves itself by using onboard sensors (accelerometers, torque sensors) to monitor landing progression and autonomously modify control parameters without requiring external high-accuracy trajectory planning
Solution Approach 2:
The patent changes control parameters (joint torques, motion trajectories) in real-time based on detected landing motion states. By dynamically adjusting these parameters during the landing process rather than relying on fixed high-accuracy pre-planned curves, the system achieves effective impact reduction with simpler, more cost-effective hardware
Data Source
AI summary
A footed robot landing control method and device are provided. The footed robot landing control method includes: detecting a landing motion state of the robot; if the landing motion state is a flight phase descending state, a motion of the foot portion of the robot with respect to a ground in the flight phase descending state is controlled based on a relative speed; if the landing motion state is a support phase landing state, a motion of joints of the robot in the support phase landing state is controlled based on a first expected joint torque. The footed robot landing control method and device are capable of reducing the impact of the foot portion against the ground, thereby realizing the flexible control of the landing process of the footed robot in a simple and rapid manner and reducing the cost of the footed robot.


