Bipedal Robot Walking Control via Capture Point and Hip Angle
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing robot walking technologies face challenges in achieving stable and natural three-dimensional walking while stretching knees, leading to reduced stride and increased energy consumption, and are limited in simulating human-like walking efficiency and stability.
Innovation Solution
A robot with multiple legs and an upper body, equipped with a sensor unit, kinematics calculation unit, target setting unit, compensation force calculation unit, virtual gravity setting unit, and servo control unit, which calculates and applies gravity compensation torques to maintain balance and stability, allowing for natural and efficient three-dimensional walking.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the robot walks while bending the knees, then the inverse kinematics can be calculated, but the stride is reduced and energy consumption is increased
Solution Approach 1:
The patent changes the control parameter from knee angle (bent knee configuration) to hip angle (straight knee configuration). By controlling the hip joint angle instead of bending the knee, the robot achieves stable walking while maintaining straight knees, thereby increasing stride length and reducing energy consumption associated with continuous knee flexion and extension.
2Ease of operation
If the robot implements three-dimensional walking, then the walking is more natural like human, but the analysis becomes complicated and stability is reduced
Solution Approach 1:
The patent segments the walking control into independent joint control tasks. Instead of analyzing complex three-dimensional whole-body dynamics, the control is divided into controlling each joint (hip, knee, ankle) independently based on desired trajectory. This segmentation simplifies the control analysis while maintaining natural three-dimensional walking capability.
Solution Approach 2:
The patent implements feedback control by continuously measuring the actual joint angles and comparing them with the desired angles, then adjusting the motor torques accordingly. This feedback mechanism ensures stability during natural three-dimensional walking by correcting deviations in real-time without requiring complex forward dynamics analysis.
3Reliability
If the robot controls joint angles using inverse kinematics, then the ZMP can be maintained within support polygon, but the knees must be bent which reduces walking efficiency
Solution Approach 1:
The patent transitions from static inverse kinematics control to dynamic trajectory-based control. Instead of calculating fixed joint angles from inverse kinematics, the system defines desired trajectories for each joint that naturally maintain ZMP within the support polygon while allowing straight knee configuration, thereby improving walking efficiency through dynamic motion planning.
Data Source
AI summary
A bipedal robot having a pair of legs with 6 degrees of freedom and a control method thereof which calculate a capture point by combining the position and velocity of the center of gravity (COG) and control the capture point during walking to stably control walking of the robot. A Finite State Machine (FSM) is configured to execute a motion similar to walking of a human, and thus the robot naturally walks without constraint that the knees be bent all the time, thereby being capable of walking with a large stride and effectively using energy required while walking.