Biped Robot Footprint Generation for Collision-Free Step Planning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for designing and selecting the size of footprint sets for biped robots in complex environments are inefficient, leading to a high probability of no solution and reduced navigation planning efficiency when the preset footprint set is too large or too small.
Innovation Solution
A dynamic footprint set generation method for biped robots, which involves obtaining preset calculation parameters, calculating landing point displacements and rotations, performing collision detection, and adjusting displacement angles to generate a set of feasible landing points, thereby improving navigation planning efficiency and reducing the probability of no solution.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the preset footprint set is too large, then the probability of finding a solution increases, but the efficiency of navigation planning drops sharply
Solution Approach 1:
The patent implements a dynamic footprint set generation method that adjusts the footprint set size based on environmental complexity and navigation progress. Instead of using a fixed large footprint set, the system dynamically generates appropriate footprint sets during navigation planning, adapting the set size to current needs and reducing computational burden while maintaining solution reliability.
Solution Approach 2:
The system changes key parameters (footprint set size, displacement angle range, collision detection thresholds) dynamically during the navigation process. By adjusting these parameters based on environmental conditions and planning stage, the system optimizes the balance between finding solutions and maintaining planning efficiency.
2Productivity
If the preset footprint set is too small, then the efficiency of navigation planning improves, but the probability of no solution increases
Solution Approach 1:
The patent performs preliminary collision detection and footprint validation before final navigation planning. By pre-processing and filtering feasible footprints early in the process, the system reduces the need for large footprint sets while ensuring that sufficient valid options remain for finding navigation solutions.
Solution Approach 2:
The system incorporates feedback mechanisms where collision detection results and planning outcomes inform subsequent footprint set generation. This feedback loop allows the system to learn from previous planning attempts and adjust footprint set characteristics to maintain solution probability while optimizing efficiency.
3Speed
If the displacement angle adjustment amplitude is large, then the search speed for landing points increases, but the precision of landing point selection decreases
Solution Approach 1:
The patent employs periodic adjustment of displacement angle amplitudes during the search process. The system uses larger amplitude adjustments in early search phases to quickly explore the solution space, then transitions to smaller amplitude adjustments as the search converges, thereby achieving both fast initial search and precise final selection.
Solution Approach 2:
The displacement angle adjustment amplitude is made dynamic rather than fixed. The system adapts the amplitude based on search progress, environmental complexity, and convergence status, allowing it to optimize the trade-off between search speed and precision at different stages of the navigation planning process.
Data Source
AI summary
A dynamic footprint set generation method, a biped robot using die same, and a computer readable storage medium are provided. The method includes: obtaining preset footprint calculation parameters; calculating a landing point position based on the preset footprint calculation parameters; determining a landing point range based on a landing point position, and performing a collision detection on the landing point range; recording the corresponding landing point position in a footprint set in response to the detection result representing there being no collision; obtaining a preset adjustment amplitude to update a preset displacement angle after the recording is completed; and returning to the calculating the landing point position until the footprint set is generated. By continuously adjusting the preset displacement angle, each landing point position is calculated accordingly, and the valid landing point positions are recorded in the footprint set, which provides more feasible landing points for navigation planning.