3D Depth Camera Navigation for Close-Range Robot Obstacle Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous indoor robots face challenges in navigation due to obstacles at various altitudes and the distance between obstacle sensors and the floor, especially in unstructured environments with untrained or inattentive personnel, requiring refined obstacle handling for safety reasons.
Innovation Solution
The use of two three-dimensional depth camera sensors mounted on an autonomous robot, capable of rotation for a 360-degree field of view, with a frame rate of 15-90fps, to provide accurate obstacle detection and navigation by comparing depth images to templates for precise path planning and safety monitoring.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional obstacle sensors are used on autonomous robots, then the robot can detect obstacles, but the detection accuracy is insufficient for obstacles at various altitudes and distances
Solution Approach 1:
The patent transitions from conventional 2D depth sensing to 3D depth camera sensors that capture depth information in three-dimensional space. This dimensional enhancement allows the robot to detect obstacles at various altitudes and distances with higher precision, resolving the measurement accuracy issue while maintaining manageable system complexity through sophisticated processing of the additional dimensional data.
Solution Approach 2:
The patent employs multiple depth camera sensors positioned at different locations on the robot body, each capturing depth information from specific spatial segments. This segmentation of the sensing field allows comprehensive coverage of obstacles at various altitudes and distances, improving overall detection accuracy without requiring a single overly complex sensor system.
2Adaptability or versatility
If the robot operates in unstructured environments with untrained personnel, then the robot gains operational versatility, but safety and obstacle handling become more difficult
Solution Approach 1:
The patent creates depth image templates representing safe travel paths before the robot begins navigation in unstructured environments. These templates serve as pre-established reference models that enable the robot to reliably identify and avoid obstacles, including those posed by untrained personnel, thereby maintaining safety and reliability while operating in versatile, unstructured settings.
Solution Approach 2:
The patent implements a feedback mechanism where captured depth images are continuously compared against templates to identify deviations indicating obstacles. This real-time feedback allows the robot to adapt its navigation in unstructured environments while maintaining high safety standards, as the system continuously monitors and responds to environmental changes including unexpected obstacles.
3Area of stationary object
If depth camera sensors are positioned higher on the robot body to increase field of view, then the detection range is improved, but the distance to the floor increases making obstacle detection more difficult
Solution Approach 1:
The patent uses 3D depth camera sensors that capture comprehensive three-dimensional depth information, allowing the system to accurately measure distances to the floor even from elevated positions. The three-dimensional capability compensates for the increased height, maintaining measurement precision while enjoying the expanded field of view benefits of higher sensor placement.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An apparatus, system and method of operating an autonomous mobile robot having a height of at least one meter. The robot body; at least two three-dimensional depth camera sensors affixed to the robot body proximate to the height, wherein the sensors are directed toward a floor surface and, in combination, comprise a substantially 360 degree field of view of the floor surface around the robot body; and a processing system for receiving pixel data within the field of view of the sensors; obtaining missing or erroneous pixels from the pixel data; comparing the missing or erroneous pixels to a template, wherein the template comprises at least an indication of ones of the missing or erroneous pixels indicative of the robot body and a shadow of the robot body; and outputting an indication of obstacles in or near the field of view based on the comparing.