UAV Delivery Point Segmentation for Obstacle-Aware Abort Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Unmanned aerial vehicles (UAVs) face challenges in safely delivering payloads due to obstacles such as trees, roads, and vehicles, which can cause damage to the UAV, payload, or obstacles if delivery is attempted too close to them.
Innovation Solution
UAVs use image segmentation and semantic classification to determine a 'nudged' delivery point further away from obstacles, adjusting their flight path to avoid collisions by navigating to a safer delivery location and aborting the delivery process if the surrounding area is too obstructed.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If the UAV delivers the payload close to the target location, then the delivery precision is improved, but the risk of collision with obstacles increases
Solution Approach 1:
The system performs preliminary detection of obstacles using sensors and image segmentation before the UAV reaches the delivery location. This advance detection allows the UAV to identify hazardous areas and adjust its delivery point accordingly, maintaining precision while avoiding collisions
Solution Approach 2:
The system introduces an intermediary processing layer between the target location and the actual delivery point. This layer uses semantic segmentation to analyze the environment and determine a safe delivery point that is offset from the original target if obstacles are detected, thus mediating between precision requirements and collision avoidance
2Reliability
If the UAV uses semantic segmentation to identify obstacles, then the collision avoidance capability is improved, but the computational complexity increases
Solution Approach 1:
The system divides the image processing task into semantic segmentation of the environment into distinct regions (obstacles, safe zones, delivery areas). This segmentation allows the UAV to efficiently identify obstacle pixels and calculate the percentage of obstacle pixels in the surrounding area without processing the entire image at full complexity
Solution Approach 2:
The system changes the parameter of interest from analyzing every pixel in detail to calculating the percentage of obstacle pixels in a defined surrounding area. This parameter transformation simplifies the computational burden while maintaining reliable collision detection by focusing on the critical metric of obstacle density in the delivery zone
Data Source
AI summary
A method includes capturing, by a sensor on an unmanned aerial vehicle (UAV), an image of a delivery location. The method also includes determining, based on the image of the delivery location, a segmentation image. The segmentation image segments the delivery location into a plurality of pixel areas with corresponding semantic classifications. The method additionally includes determining, based on the segmentation image, a percentage of obstacle pixels within a surrounding area of a delivery point at the delivery location, wherein each obstacle pixel has a semantic classification indicative of an obstacle in the delivery location. The method further includes based on the percentage of obstacle pixels being above a threshold percentage, aborting a delivery process of the UAV.