Humanoid Robot Leg Control via Human Posture Mapping

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing humanoid robots have low adaptability and stability on complex terrains due to their fixed programming methods, which are not flexible enough to handle varied environments effectively.

Innovation Solution

A humanoid robot control method that collects posture information from human leg joints using MEMS inertial sensors, maps this information to leg joint servos, and optimizes the rotation angles and angular velocities based on the stability theory of the extrapolated centroid XCoM to improve stability and flexibility, allowing the robot to maintain balance and perform more complex movements without relying on precise dynamic models.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If fixed programming method is used for humanoid robot control, then the control system is simple and easy to implement, but the adaptability to complex terrains is low

Engineering Contradiction:

Improveadaptability to complex terrainsVSAvoidcontrol system complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The humanoid robot uses its own inertial sensors to collect posture information and automatically maps this information to joint servos through the control system. The robot self-adjusts its movements based on real-time sensor data without requiring external intervention or complex pre-programming for each terrain condition

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The control system dynamically adjusts joint servo parameters (rotation angles and angular velocities) based on real-time posture information from inertial sensors. The system transitions from fixed programming to dynamic adaptation by continuously optimizing movement parameters according to actual terrain conditions and robot state

Inventive Principle:

Principle #15Dynamics

2Measurement precision

If precise dynamic model is used for gait planning, then the movement accuracy is improved, but the computational complexity and processing time increase

Engineering Contradiction:

Improvemovement accuracyVSAvoidcomputational processing time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

Instead of using complex precise dynamic models for gait planning, the system copies human movement patterns by mapping posture information from inertial sensors directly to joint servo parameters. This approach replicates natural human gait characteristics without requiring computationally intensive dynamic simulations

Inventive Principle:

Principle #26Copying

Solution Approach 2:

The patent replaces traditional mechanical gait planning based on precise dynamic models with a sensor-based feedback system. Inertial sensors collect real-time posture data, and the control system directly translates this information into joint commands, substituting complex mechanical calculations with real-time sensing and control

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

3Reliability

If traditional control method is used, then the system structure is simple, but the stability on complex terrains is low

Engineering Contradiction:

Improvestability on complex terrainsVSAvoidcontrol system structure

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The control system continuously collects posture information from inertial sensors and uses this feedback to adjust joint servo parameters in real-time. This closed-loop feedback mechanism enables the robot to maintain stability on complex terrains by constantly adapting to changing conditions based on actual sensor data

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The system performs preliminary mapping of posture information to joint servo parameters before actual movement execution. By pre-processing sensor data and establishing the relationship between posture and joint commands in advance, the system prepares optimal movement parameters proactively, improving stability response

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS11904472B2Humanoid robot control method, humanoid robot using the same, and computer readable storage medium

Publication Date: 2024.02.20 UBTECH ROBOTICS CORP LTD

AI summary

A humanoid robot control method, a mobile machine using the same, and a computer readable storage medium are provided. The method includes: mapping posture information of leg joints of a human body to leg joint servos of a humanoid robot to obtain an expected rotation angle and an expected rotation angular velocity of non-target optimized joint servos of the leg joint servos and an expected rotation angle and an expected rotation angular velocity of target optimized joint servos of the leg joint servos; obtaining an optimization objective function corresponding to the target optimized joint servos of the leg joint servos; optimizing the expected rotation angle and the expected rotation angular velocity of the target optimized joint servos to obtain a corrected expected rotation angle and a corrected expected rotation angular velocity of the target optimized joint servos; and controlling each of the leg joint servos of the humanoid robot.