Design and Development of a Real-time Multi-object Detection and Tracking Vision System for MP-R Patrol Robot**********

MP-R巡逻机器人实时多目标检测与跟踪视觉系统的设计与开发************

• 批准号: 536008-2018
• 负责人: Lang, Haoxiang
• 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=656571
• 关键词: Design; Development; Real; time; Multi

Public safety has become one of the challenges nowadays since the potential threats of modern cities have dramatically changed. Moreover, this issue highly relates to the attraction of people and the success of economy. Public safety usually relies on the ability of the local and federal police. However, there is always a limitation of coverage which may lead to serious tragedy due to crime and terrorism. Therefore, patrolling robots can be involved to address the issue.**The objective of this proposal presents a six-month research and development plan of a real-time object identification and tracking system based on the MR-P robot platform developed by ESI's Robotic Technology. The project will address challenges and have abilities: 1) Real-time vision and image processing system; 2) Robust object identification and tracking and 3) Pose estimation of the tracked object.**The outcomes of the project and its applications will greatly improve the current vision capability of the robot, which can better serve the product goal of the ESI.**

公共安全已成为当今的挑战之一，因为现代城市的潜在威胁发生了巨大变化。此外，这一问题与人口的吸引力和经济的成功密切相关。公共安全通常依赖于当地和联邦警察的能力。然而，由于犯罪和恐怖主义，报道总是有限的，这可能导致严重的悲剧。因此，巡逻机器人可以参与解决这个问题。本提案的目的是提出一个为期六个月的研究和开发计划的实时对象识别和跟踪系统的基础上的MR-P机器人平台开发的ESI的机器人技术。该项目将解决挑战并具有以下能力：1）实时视觉和图像处理系统; 2）鲁棒的对象识别和跟踪; 3）跟踪对象的姿态估计。该项目的成果及其应用将大大提高机器人目前的视觉能力，从而更好地服务于ESI的产品目标。