Humanoid Robot Balance via Center of Mass Trajectory
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Humanoid robots struggle to perform actions like standing up from a seated position without falling due to differences in kinematic and dynamic properties between humans and robots, making it difficult to replicate human movements accurately.
Innovation Solution
The approach involves receiving motion capture data and contact force data from human demonstrations to approximate the center of mass (CoM) trajectory, then generating a planned robot action using inverse kinematics with the approximated CoM trajectory and motion capture data as constraints, allowing the robot to emulate human-like movements and maintain balance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If humanoid robots replicate human movements directly, then the robot can perform anthropomorphic actions, but the robot loses balance and falls over
Solution Approach 1:
The system changes the parameters of the movement by approximating the center of mass trajectory rather than directly copying human motion. The inverse kinematics solver adjusts joint angles to follow the approximated CoM path, adapting the motion parameters to the robot's specific kinematic and dynamic properties while maintaining the essential characteristics of the human demonstration
Solution Approach 2:
The approximated center of mass trajectory serves as an intermediary between the human demonstration and the robot execution. Instead of directly transferring human motion data, the system uses the CoM trajectory as a mediator that can be adapted to the robot's properties through inverse kinematics, enabling balance maintenance while preserving anthropomorphic action capabilities
2Stability of the object's composition
If humanoid robots are bolted to the floor, then the robot can maintain balance, but the robot cannot move freely
Solution Approach 1:
The robot achieves balance through self-control of its center of mass trajectory rather than external anchoring. The inverse kinematics system continuously adjusts the robot's joint configurations to maintain the approximated CoM path, enabling the robot to stabilize itself dynamically during movement without requiring mechanical attachment to the environment
3Manufacturing precision
If motion capture data is directly transferred to the robot, then the robot can replicate human actions, but the robot's different kinematic properties cause balance loss
Solution Approach 1:
Instead of directly transferring motion capture data to robot joints, the system inverts the approach by first extracting the center of mass trajectory from the demonstration and then using inverse kinematics to compute the joint angles. This inversion allows the system to prioritize balance (CoM control) over exact motion replication, resolving the conflict between action accuracy and stability
Data Source
AI summary
Embodiments of the invention provide an approach for reproducing a human action with a robot. The approach includes receiving data representing motions and contact forces of the human as the human performs the action. The approach further includes approximating, based on the motions and contact forces data, the center of mass (CoM) trajectory of the human in performing the action. Finally, the approach includes generating a planned robot action for emulating the designated action by solving an inverse kinematics problem having the approximated human CoM trajectory as a hard constraint and the motion capture data as a soft constraint.