Bipedal Robot Gait Generation Using Dual Dynamic Models
Solution Overview
Problem
Existing technologies for generating desired gaits in bipedal robots fail to prevent slipping and maintain dynamic accuracy, especially on low-friction surfaces or during motions with zero translational floor reaction force, due to neglecting translational floor reaction force errors and high nonlinearity in dynamic models.
Innovation Solution
A gait producing device that uses a first dynamic model to create a provisional motion and corrects it using a second dynamic model with higher accuracy, ensuring the desired floor reaction force moment and permissible range are satisfied by adjusting the provisional motion based on errors in floor reaction force moment and translational force components.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a gait is generated using conventional technologies that only consider floor reaction force moment errors, then the generation process is simple, but the robot slips on low-friction surfaces or during running motions
Solution Approach 1:
The gait generation process is segmented into two distinct stages: a first dynamic model stage that generates provisional motion with basic stability, and a second dynamic model stage that corrects the motion to prevent slipping. This segmentation allows each stage to focus on specific requirements without overwhelming complexity.
Solution Approach 2:
The first dynamic model performs preliminary action by generating a provisional motion that satisfies basic dynamic balance conditions. This preliminary gait serves as the foundation that is subsequently refined by the second dynamic model to achieve slip prevention.
2Measurement precision
If a single high-accuracy dynamic model is used for gait generation, then dynamic accuracy is improved, but the computational efficiency and real-time performance deteriorate
Solution Approach 1:
The dynamic model is segmented into two parts: a first dynamic model with simplified relationships that enables fast computation, and a second dynamic model with high accuracy that provides precise correction. This segmentation resolves the conflict between speed and accuracy.
Solution Approach 2:
The first dynamic model acts as an intermediary that generates a provisional motion quickly. This intermediate result is then refined by the second dynamic model, allowing the system to benefit from both fast preliminary generation and accurate final correction.
3Ease of operation
If the second dynamic model corrects arbitrary instantaneous desired motion using past values, then temporal coordination is simplified, but dynamic accuracy deteriorates due to temporal differences
Solution Approach 1:
The correction process uses feedback from the second dynamic model by calculating the difference between the actual floor reaction force moment and the desired floor reaction force moment. This feedback mechanism ensures temporal coordination while maintaining dynamic accuracy.
Solution Approach 2:
The first dynamic model performs preliminary action by generating a provisional motion that is close to the desired motion. This reduces the magnitude of correction needed by the second dynamic model, improving both efficiency and accuracy.
Data Source
AI summary
By using a first dynamic model of a moving robot 1, a provisional motion, which indicates a provisional value of a desired motion of the robot 1, is created such that a desired value of a floor reaction force moment horizontal component and a permissible range of a translational floor reaction force horizontal component are satisfied on the first dynamic model. The difference between a floor reaction force produced on a second dynamic model, which has a dynamic accuracy that is higher than that of the first dynamic model, by the provisional motion and a floor reaction force produced on the first dynamic model is defined as a floor reaction force error. Based on this floor reaction force error, the provisional motion is corrected on the first dynamic model to generate a desired motion. The desired motion is generated such that the value obtained by adding the floor reaction force error to the floor reaction force generated on the first dynamic model satisfies the aforesaid desired value and permissible range.