Object detection and hazard avoidance with FPV equipped single-rotor UAVs

使用配备 FPV 的单旋翼无人机进行物体检测和危险规避

• 批准号: 530001-2018
• 负责人: Azim, Akramul
• 金额: $ 1.82万
• 依托单位: University of Ontario Institute of Technology
• 依托单位国家: 加拿大
• 项目类别: Engage Grants Program
• 财政年份: 2018
• 资助国家: 加拿大
• 起止时间: 2018-01-01 至 2019-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=644382
• 关键词: Object; detection; hazard; avoidance; FPV

Design challenges exist for obstacle detection and hazard avoidance of unmanned aerial vehicle (UAV)**systems. Our research addresses this challenge by analyzing tools, techniques and architectures that assist**developers in building robust and reliable UAVs equipped with efficient obstacle detection and hazard**avoidance methods. NOVAerial Robotics is a manufacturer of UAV and a well-received one is Procyon 800E**which a single rotor helicopter style UAV. One of the challenges in operating UAVs by NOVAerial robotics is**the detection of objects in different environmental conditions. Therefore, in this project we aim to investigate**and propose solutions to address the problem of object detection and hazard avoidance in first person view**(FPV) equipped single rotor UAVs.**UAVs need to provide reliable service with appropriate adaptation in various dynamic situations. The system**needs to keep track of all the situations for proper functioning. It is necessary for the system to have sufficient**knowledge of its different conditions and properties by monitoring its environment. Recent advances in**imaging techniques and communications have introduced an incredible growth of different cameras and**various visual devices as a powerful monitoring tool. Moreover, video provides detailed information on the**interactions as opposed to tracing or data logging. Through analyzing video data, we can detect different events**and identify their interactions including timing properties. This project aims to analyze video streams gathered**from operating UAVs for obstacle detection and hazard avoidance.

障碍物检测和危害无人机（UAV）**系统存在设计挑战。我们的研究通过分析工具，技术和体系结构来解决这一挑战，这些工具，技术和架构可以帮助**开发人员建立具有高效障碍物检测和危害的可靠无人机**避免方法。 Novaerial Robotics是无人机的制造商，而广受好评的是Procyon 800E **，它是单个转子直升机风格的无人机。通过新闻机器人技术操作无人机的挑战之一是**在不同环境条件下检测物体。因此，在这个项目中，我们旨在调查**并提出解决方案，以解决第一人称视图中的对象检测和避免危险的问题**（FPV）配备的单转子无人机。系统**需要跟踪所有情况以进行正确的功能。系统必须通过监视其环境来了解其不同条件和特性的足够**知识。 **成像技术和通信的最新进展引入了不同的相机和**各种视觉设备作为强大的监视工具的令人难以置信的增长。此外，视频提供了有关**交互的详细信息，而不是跟踪或数据记录。通过分析视频数据，我们可以检测到不同的事件**并确定其相互作用，包括正时属性。该项目旨在分析收集的视频流，从操作无人机进行障碍物检测和避免危害。