UAV Delivery Abort Using Semantic Obstacle Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Unmanned aerial vehicles (UAVs) face challenges in safely delivering payloads due to obstacles in the delivery location, which can result in collisions and damage to the UAV, payload, and obstacles.
Innovation Solution
The UAV captures an image of the delivery location, determines a segmentation image to identify pixel areas with obstacles, calculates the percentage of obstacle pixels in the surrounding area, and aborts the delivery process if the percentage exceeds a threshold.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the UAV delivers the payload at the initial delivery point, then the delivery efficiency is improved, but the risk of collision with obstacles increases
Solution Approach 1:
The system performs preliminary action by capturing an image of the delivery location and determining a segmentation image before the UAV reaches the delivery point. This advance analysis of the environment allows the system to identify obstacles and calculate the percentage of obstacle pixels in the surrounding area, enabling proactive adjustment of the delivery point to avoid collisions while maintaining delivery efficiency
Solution Approach 2:
The system implements feedback by continuously monitoring the segmentation image and obstacle pixel percentage, then adjusting the delivery point based on this information. If the percentage of obstacle pixels exceeds a threshold, the system aborts the delivery process or selects an alternative delivery point, ensuring safe operation while maintaining productivity
2Reliability
If the UAV selects a nudged delivery point further from obstacles, then the safety is improved, but the delivery precision to the original location deteriorates
Solution Approach 1:
The system applies local quality by making the delivery point adjustment localized rather than global. Instead of moving the delivery point far from the original location, the system nudges it only as much as necessary to avoid obstacles, based on the calculated percentage of obstacle pixels in the surrounding area. This ensures safety while minimizing the impact on delivery precision
Solution Approach 2:
The system changes parameters by adjusting the delivery point coordinates based on the segmentation image analysis. The adjustment magnitude is controlled by the obstacle pixel percentage threshold, allowing the system to optimize between safety and precision by modifying the delivery point parameters dynamically according to environmental conditions
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.