UAV Obstacle Avoidance Using Invalid Stereo Depth Pixels
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Unmanned aerial vehicles (UAVs) face challenges in navigating through environments where certain objects, such as power lines, cannot be accurately measured by stereo cameras, posing hazards during descent for delivery operations.
Innovation Solution
Implementing an image processing system that receives depth images from stereo cameras, identifies pixels with invalid depth values, and infers the presence of potential obstacles, generating an invalid pixel mask to control the UAV's movement and avoid obstacles, particularly using a Hough transform to detect elongated objects like power lines.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If stereo camera is used for depth measurement, then navigation capability is improved, but measurement precision deteriorates for certain objects like power lines
Solution Approach 1:
The patent introduces an intermediary processing layer that analyzes the depth image to identify invalid depth values and generates an invalid depth mask. This mask serves as a mediator between the stereo camera's limited measurement capability and the navigation system, allowing the UAV to compensate for the camera's inability to measure certain objects by using alternative detection methods and obstacle avoidance algorithms.
2Reliability
If obstacle avoidance system is added, then safety is improved, but device complexity increases
Solution Approach 1:
The patent makes the image processing system multi-functional by using the same depth image processing pipeline for both navigation and obstacle detection. The invalid depth mask generation and analysis routines serve dual purposes: they identify measurement limitations for navigation planning while simultaneously detecting potential obstacles. This universal approach improves safety without requiring entirely separate dedicated obstacle avoidance hardware.
3Reliability
If invalid depth values are identified and processed, then obstacle detection is improved, but processing time increases
Solution Approach 1:
The patent extracts only the critical information needed for obstacle detection by generating an invalid depth mask that isolates pixels with unreliable depth values. Rather than processing the entire depth image in detail, the system extracts the specific regions of interest (invalid depth areas) and focuses computational resources on analyzing these extracted regions for obstacle patterns, thereby reducing overall processing time while maintaining detection accuracy.
Data Source
AI summary
A computer-implemented method comprises receiving, by an image processing system, a depth image captured by a stereo camera on an unmanned aerial vehicle (UAV). One or more pixels of the depth image are associated with corresponding depth values indicative of distances of one or more objects to the stereo camera. The image processing system determines that one or more pixels of the depth image are associated with invalid depth values. The image processing system infers, based on a distribution of the one or more pixels of the depth image that are associated with invalid depth values, a presence of a potential obstacle in an environment of the UAV. The UAV is controlled based on the inferred presence of the potential obstacle.