Robot Footfall Clustering for Stair Geometry Identification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Robots lack the ability to precisely navigate stairs and stair-like terrain due to the absence of natural coordination, which can lead to missteps, slips, or falls, highlighting the need for systems and methods to coordinate precise leg movements.
Innovation Solution
A method and system that identify stairs from footfalls by clustering footfall locations based on height, determining candidate pairs that satisfy stair height and depth thresholds, and generating a stair model to orient the robot for safe navigation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If robots traverse stairs without precise coordination, then navigation speed is maintained, but safety and reliability deteriorate due to missteps, slips, or falls
Solution Approach 1:
The patent segments the stair navigation task into distinct phases: footfall detection, candidate pair identification, clustering analysis, and stair model generation. Each phase processes specific data independently, allowing the system to manage complexity through modular analysis rather than requiring holistic coordination of all leg movements simultaneously.
Solution Approach 2:
The system performs preliminary actions by pre-identifying stair geometries and generating stair models before actual navigation. Footfall locations are collected and analyzed in advance to create a predictive stair model, allowing the robot to plan its leg movements based on preprocessed environmental information rather than reacting in real-time during traversal.
2Reliability
If robots use complex coordination systems for stair navigation, then navigation safety improves, but processing time and computational resources increase
Solution Approach 1:
The patent transforms the navigation problem by changing parameters from continuous spatial coordinates to discrete clustered groups. By converting footfall locations into height-based clusters and identifying candidate pairs through parameter comparison against thresholds, the system reduces computational complexity and enables faster processing while maintaining accurate stair geometry detection.
Solution Approach 2:
The system creates a simplified copy of the physical stair structure as a digital stair model. This virtual representation captures essential geometric features (riser heights, tread depths, orientation) without requiring full physical complexity, allowing rapid analysis and navigation planning based on the model rather than processing raw sensor data in real-time.
3Measurement precision
If robots accurately identify and orient to stair models, then navigation precision improves, but the complexity of environmental perception systems increases
Solution Approach 1:
The robot's own footfalls serve as the sensing mechanism for stair detection. By using its operational movements (leg contacts with surfaces) to generate the data needed for stair identification, the system eliminates the need for separate, complex perception hardware. The navigation system is self-informing through the robot's intrinsic interaction with the environment during traversal.
Solution Approach 2:
The footfall detection system serves multiple functions: it tracks robot position for navigation, identifies stair geometries for obstacle detection, and provides data for generating the stair model. This multi-functionality reduces the need for dedicated specialized sensors and processing systems, simplifying the overall perception architecture while maintaining high measurement precision.
Data Source
AI summary
A method of identifying stairs from footfalls includes receiving a plurality of footfall locations of a robot traversing an environment. Each respective footfall location indicates a location where a leg of the robot contacted a support surface. The method also includes determining a plurality of candidate footfall location pairs based on the plurality of footfall locations. The candidate footfall location pair includes a first and a second candidate footfall location. The method further includes clustering the first candidate footfall location into a first cluster group based on a height of the first candidate footfall location and clustering the second candidate footfall location into a second cluster group based on a height of the second candidate footfall location. The method additionally includes generating a stair model by representing each of the cluster groups as a corresponding stair and delineating each stair based on a respective midpoint between each adjacent cluster group.