Adjustable Stereo Camera Scanning for Robot Blind Spot Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Robots face uncertainty and increased collision risks due to unknown or inadequately determined environmental objects and attributes, particularly in areas with blind spots or limited vision component coverage, which can lead to suboptimal operation and increased chances of collisions.
Innovation Solution
Implementing a system with multiple vision components, including a fixed LIDAR and an adjustable stereo camera, dynamically adjusting the camera's pose to address blind spots and enhance data accuracy, combined with adjustable robot arms to avoid obstructing the camera's view, thereby improving object detection and attribute determination.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a fixed vision component is used, then the device complexity is reduced, but the measurement precision and coverage of environmental objects are insufficient due to blind spots
Solution Approach 1:
The patent applies the dynamics principle by making the vision component adjustable rather than fixed. The vision component can dynamically change its pose (position and orientation) to observe different areas along the robot's predicted trajectory. This allows the system to eliminate blind spots and improve detection accuracy without requiring multiple fixed vision components, thus resolving the contradiction between measurement precision and device complexity.
2Reliability
If the robot reduces velocity to account for unknown conditions, then the reliability is improved, but the productivity decreases
Solution Approach 1:
The patent applies preliminary action by proactively adjusting the vision component to observe potential blind spots and unknown areas before the robot reaches them. By capturing vision data in advance along the predicted trajectory, the system identifies environmental objects and conditions beforehand, allowing the robot to maintain higher velocities while still ensuring collision avoidance through prior knowledge of the environment.
3Measurement precision
If multiple vision components are used, then the measurement precision is improved, but the device complexity increases
Solution Approach 1:
Instead of using multiple fixed vision components simultaneously, the patent employs a single vision component that can dynamically adjust its pose to capture data from multiple perspectives along the robot's trajectory. This dynamic approach achieves comprehensive environmental coverage and high detection accuracy equivalent to or better than multiple fixed components, while significantly reducing device complexity.
4Loss of information
If the adjustable vision component is used to cover blind spots, then the loss of information is reduced, but the ease of operation decreases due to coordination requirements
Solution Approach 1:
The patent implements feedback mechanisms where the robot's predicted trajectory and current vision data are continuously monitored. Based on this feedback, the system automatically determines which areas require additional observation and adjusts the vision component's pose accordingly. This closed-loop control approach minimizes information loss by proactively identifying and observing blind spots while automating the coordination process, thereby reducing the operational burden on the system.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach reduces uncertainty and enhances the accuracy of environmental object detection, leading to more optimal robot navigation and reduced collision risks by effectively mitigating blind spots and improving data coverage.
Implementation Method 1
a fixed LIDAR
Implementation Method 2
an adjustable stereo camera
Data Source
Figure 1
Figure 2A
Figure 2B
AI summary
Implementations set forth herein relate to a robot that employs a stereo camera and LIDAR for generating point cloud data while the robot is traversing an area. The point cloud data can characterize spaces within the area as occupied, unoccupied, or uncategorized. For instance, an uncategorized space can refer to a point in three-dimensional (3D) space where occupancy of the space is unknown and/or where no observation has been made by the robot—such as in circumstances where a blind spot is located at or near a base of the robot. In order to efficiently traverse certain areas, the robot can estimate resource costs of either sweeping the stereo camera indiscriminately between spaces and/or specifically focusing the stereo camera on uncategorized space(s) during the route. Based on such resource cost estimations, the robot can adaptively maneuver the stereo camera during routes while also minimizing resource consumption by the robot.