Uncertainty-aware full-body motion planning of aerial and multi-legged robots for urban search and rescue operations

用于城市搜救行动的空中和多足机器人的不确定性全身运动规划

• 批准号: 560791-2020
• 负责人: RamirezSerrano, Alejandro
• 金额: $ 7.29万
• 依托单位: University of Calgary
• 依托单位国家: 加拿大
• 项目类别: Alliance Grants
• 财政年份: 2021
• 资助国家: 加拿大
• 起止时间: 2021-01-01 至 2022-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=720674
• 关键词: Uncertainty; aware; full; body; motion

With the increasing severity and frequency of natural disasters such as tornados, floods, and many others, the 21st Century presents unique challenges to disaster response. The capabilities of autonomous robots have the potential to be used as effective tools for Urban Search & Rescue (USAR) operations in complex confined environments - both structured and unstructured - such as those encountered inside collapsed buildings and in burning infrastructure, among many others. However, major challenges related to autonomy, locomotion, and adaptability remain before robots can be effectively deployed inside such spaces where safety-critical decisions must often be made despite uncertainty in the estimation, execution, and the environment. These problems can often be cast as a class of problems for which finding an exact solution is challenging. This project will develop whole-body tractable motion-planning solutions enabling robotic systems to operate inside chaotic hazardous confined spaces where real-time operation is essential. The results will enable to deploy drones and multi-legged robots under uncertainty estimation via sophisticated estimation schemes like visual-inertial odometry that depend on the distribution of the features in the environment. The work is important to reduce the effects of natural and other disasters that cause billions of dollars in economic losses every year. Using the wealth of experience in Canada's emergency agencies and disaster response, the project will develop effective robotic tools for USAR that can be in position and deployed directly following an incident of structural collapse. The results will render emergency preparedness and response activities more effective and thereby save more lives and reduce community recovery time. The outcomes will improve efficiency in disaster response operations at a disaster site and provide the means to increase capacity to prepare for, mobilize and coordinate USAR assistance in support of affected communities in collapsed structure emergencies. The results will support capacity-building at the national, global, and regional levels. In collaboration with first response organizations, 6 graduate and 10 BSc students will be trained in a multidisciplinary team environment.

随着龙卷风、洪水等自然灾害日益严重和频繁，21 世纪对灾害应对提出了独特的挑战。自主机器人的功能有可能成为复杂受限环境（结构化和非结构化）中城市搜索与救援（USAR）行动的有效工具，例如在倒塌的建筑物和燃烧的基础设施等中遇到的环境。然而，在机器人能够有效部署在此类空间内之前，与自主性、运动和适应性相关的重大挑战仍然存在，尽管估计、执行和环境存在不确定性，但在这些空间中通常必须做出安全关键决策。这些问题通常可以归为一类很难找到精确解决方案的问题。该项目将开发全身易处理的运动规划解决方案，使机器人系统能够在实时操作至关重要的混乱危险密闭空间内运行。研究结果将使得能够通过复杂的估计方案（例如取决于环境中特征的分布的视觉惯性里程计）在不确定性估计下部署无人机和多足机器人。这项工作对于减少每年造成数十亿美元经济损失的自然灾害和其他灾害的影响非常重要。该项目将利用加拿大应急机构和灾难应对方面的丰富经验，为美国陆军救援队（USAR）开发有效的机器人工具，这些工具可以在结构倒塌事件发生后直接就位和部署。结果将使应急准备和响应活动更加有效，从而挽救更多生命并缩短社区恢复时间。其成果将提高灾害现场救灾行动的效率，并提供提高准备、动员和协调美国救援援助援助能力的手段，以支持在建筑物倒塌紧急情况下受影响的社区。其结果将支持国家、全球和区域各级的能力建设。与急救组织合作，6 名研究生和 10 名理学学士学生将在多学科团队环境中接受培训。