Legged Robot Barycenter Estimation From Ground Reaction Forces
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for determining the center of mass position in legged robots with a large number of joints require significant computational resources, limiting their application to robots with a small number of joints and reducing their efficiency.
Innovation Solution
A method that creates first, second, and third relation data to determine the center of mass position using a constant C, where the third relation data correlates the acting force on the leg with the center of mass position, allowing for efficient determination of the center of mass trajectory without needing to detect the state of each joint.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the center of mass position is determined by detecting the state of each joint in each leg, then the center of mass position can be calculated, but the calculation amount becomes relatively large when the legged robot has a large number of joints, limiting applicability
Solution Approach 1:
The patent extracts only the essential information needed for center of mass calculation by using force sensors on the legs to directly measure ground reaction forces, rather than detecting the state of each joint. This extraction approach obtains the necessary data (forces and moments) without the complexity of comprehensive joint state detection, especially for robots with many joints.
Solution Approach 2:
The patent introduces force sensors as intermediaries between the legged robot and the ground. These sensors directly measure the interaction forces and moments, serving as a mediator that provides the necessary information for center of mass calculation without requiring direct detection of all joint states. This intermediary approach simplifies the measurement system while maintaining accuracy.
2Measurement precision
If joint state detection is performed for each joint in each leg, then the center of mass position can be determined, but the method has narrow application range when the robot has a large number of joints
Solution Approach 1:
The patent creates a universal center of mass determination method that works for legged robots regardless of the number of joints. By using force sensors to directly measure ground reaction forces and moments, the method becomes universally applicable to various legged robot configurations (different numbers of legs, different joint arrangements) without requiring joint-specific detection systems, thereby expanding adaptability.
Solution Approach 2:
The patent extracts the essential mechanical interaction information (forces and moments at the ground contact points) that is common to all legged robots, regardless of their specific joint configurations. This extracted information serves as a universal basis for center of mass calculation that applies across different robot designs, enhancing the method's versatility.
3Measurement precision
If the center of mass position is calculated using joint state detection, then the position can be obtained, but the calculation amount is relatively large reducing movement efficiency
Solution Approach 1:
The patent uses force sensors as intermediaries that directly provide force and moment measurements, eliminating the need for complex real-time calculations based on joint states. This intermediary measurement approach reduces computational burden and enables faster center of mass determination, thereby improving movement efficiency while maintaining calculation accuracy.
Solution Approach 2:
The patent replaces the mechanical sensing approach (detecting joint states through encoders and sensors on each joint) with a force-based measurement system. By using force sensors to directly measure ground reaction forces and moments, the system substitutes complex mechanical state detection with simpler force measurements, reducing calculation requirements and improving computational efficiency.
Data Source
AI summary
A computing device creates first relation data indicating a relation between an interval duration and a center of mass position of a legged robot. The first relation data comprises a first constant, C. The computing device creates second relation data corresponding to at least one leg of for the legged robot and a force corresponding to the at last one leg with the ground. The second relation data comprises the first constant, C. The computing device creates third relation data according to the second relation data. The device determines a value of the first constant, C, when a target value J is a minimum value, and obtains the first relation data according to the determined value of the first constant, C.