Adaptive Robot Motion Control for Underactuated Equilibrium
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Controlling the motion of underactuated robots, particularly wheel-legged robots, is challenging due to the complexity of their mechanical structure, making it difficult to obtain accurate dynamic models and leading to suboptimal controller performance.
Innovation Solution
A method using adaptive dynamic programming and policy iteration to build a controller that adapts to the robot's dynamic characteristics, allowing it to maintain equilibrium and follow a target trajectory even with unknown or changing parameters, by collecting motion and control data during operation and updating the controller's parameters accordingly.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If an accurate dynamic model is designed corresponding to the mechanical structure of the robot, then the control force at each joint can be determined accurately to ensure equilibrium, but it becomes difficult to obtain an accurate dynamic model due to the complexity of the mechanical structure
Solution Approach 1:
The patent employs feedback mechanisms by continuously collecting motion state data and control data during robot operation, then using policy iteration algorithms to update the linear equilibrium parameter matrix. This closed-loop feedback system allows the controller to adapt and improve accuracy without requiring a complete accurate dynamic model upfront, resolving the contradiction between model accuracy and structural complexity.
Solution Approach 2:
The patent changes the approach from determining all dynamic model parameters accurately to iteratively updating only the linear equilibrium parameter matrix based on collected data. This parameter change strategy allows the system to achieve accurate control effects by focusing on updating specific critical parameters through policy iteration, rather than requiring complete dynamic model accuracy.
2Reliability
If the parameters of the dynamic model are known but not accurate, then the controller effects of the robot are not ideal, but obtaining accurate parameters is difficult even when the mechanical structure is known
Solution Approach 1:
The patent implements self-service by enabling the controller to automatically collect its own motion state data and control data during operation, then use policy iteration algorithms to self-update the linear equilibrium parameter matrix. This self-improving mechanism allows the system to enhance controller reliability and parameter accuracy autonomously without external intervention, resolving the contradiction between controller effectiveness and parameter accuracy.
3Stability of the object's composition
If traditional control methods are used requiring accurate dynamic models, then equilibrium can be maintained, but the design process becomes complex and requires extensive training data
Solution Approach 1:
The patent substitutes traditional mechanics-based dynamic modeling with a data-driven policy iteration approach. Instead of relying on complex mechanical structure analysis and accurate dynamic model derivation, the system uses collected motion and control data to iteratively update the linear equilibrium parameter matrix. This substitution simplifies the design process while maintaining equilibrium stability, resolving the contradiction between stability and design complexity.
Data Source
AI summary
In a method for controlling a robot, movement of the robot is controlled according to a first controller. Motion state data and control data of the robot are obtained while the movement of the robot is controlled according to the first controller. A linear equilibrium parameter matrix of the first controller is updated according to a policy iteration algorithm that uses the obtained motion state data and the control data. A second controller of the robot is built based on the updated linear equilibrium parameter matrix.