Stair Tracker Perception Using Multi-Step Edge Fitting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Robots face challenges in navigating stairs due to poor sensor data, particularly during stair ascent and descent, which can lead to inaccurate detection of edges and risers, resulting in potential missteps and collisions.
Innovation Solution
A stair tracking system that processes sensor data to detect and track features of stairs, generating a staircase model by identifying edges, walls, and heights, and refining these detections over time to improve accuracy and prevent reliance on single, potentially inaccurate readings.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the robot uses sensor data to detect stair edges, then the robot can identify stair features for navigation, but the sensor data may be inaccurate leading to poor detection reliability
Solution Approach 1:
The system performs preliminary actions by detecting edges at multiple time steps before making a final determination. The edge detection process captures candidate edges at different moments, allowing the system to pre-process and compare multiple detections before selecting the most reliable edge representation for navigation decisions.
Solution Approach 2:
The system implements feedback by comparing edge detections across multiple time steps and using this comparison to validate and refine the detected stair features. The repeated detection and comparison process creates a feedback loop that enhances both the precision and reliability of edge detection by confirming consistent features across temporal samples.
2Productivity
If the robot relies on single sensor readings for stair detection, then the processing time is reduced, but the accuracy of edge and riser detection deteriorates
Solution Approach 1:
The system applies periodic action by detecting stair edges at multiple discrete time steps rather than continuously. This periodic sampling approach allows the system to gather sufficient detection data for accurate edge identification while maintaining efficient processing by using distinct, periodic measurement moments rather than continuous monitoring.
Solution Approach 2:
The system performs preliminary detections at multiple time steps before final processing. By capturing edge candidate information in advance across several time points, the system prepares a set of potential edge locations that can be efficiently processed and compared to determine the most accurate edge representation.
3Measurement precision
If the robot detects edges at multiple time steps, then the accuracy of stair feature identification is improved, but the processing time and computational complexity increase
Solution Approach 1:
The system uses periodic action to sample edges at specific, discrete time steps rather than continuously monitoring. This approach achieves improved detection accuracy through multiple measurements while controlling processing time by using a predetermined, periodic sampling schedule that limits the total number of detections required.
Solution Approach 2:
The system performs preliminary edge detections at multiple time steps to gather accurate feature information before final processing. By collecting detection data in advance across several time points, the system prepares comprehensive edge candidate information that can be efficiently processed in a subsequent stage, separating data collection from data processing to optimize both accuracy and timing.
Data Source
AI summary
A method for perception and fitting for a stair tracker includes receiving sensor data for a robot adjacent to a staircase. For each stair of the staircase, the method includes detecting, at a first time step, an edge of a respective stair of the staircase based on the sensor data. The method also includes determining whether the detected edge is a most likely step edge candidate by comparing the detected edge from the first time step to an alternative detected edge at a second time step, the second time step occurring after the first time step. When the detected edge is the most likely step edge candidate, the method includes defining, by the data processing hardware, a height of the respective stair based on sensor data height about the detected edge. The method also includes generating a staircase model including stairs with respective edges at the respective defined heights.