Locomotion versatility robotics for hazardous industrial confined spaces operation

根据国际劳工组织的数据，全球每年大约发生230万起与工作有关的死亡事件。具体地说，在密闭空间工作时发生的事故每年造成数千人死亡和伤残。例如，艾伯塔省的石油和天然气行业报告称，每年有近3000人因密闭空间工作相关事故而致残。自主机器人有可能被用作在复杂的受限危险环境中操作的有效工具--包括结构化和非结构化的。它们在倒塌的建筑物、压力容器、筒仓、储油罐、地下电缆走廊、飞机机翼、检修孔、泵站等环境中操纵和执行危险工作的运动能力是巨大的。然而，在机器人能够有效地部署在这样的空间中之前，与自主性、运动性和适应性相关的主要挑战仍然存在，在这些空间中，尽管估计、执行和环境存在不确定性，但必须做出安全关键的决定。该项目将开发能够访问不同几何特征和危险级别的空间的新型混合移动机器人配置，并结合用于全身易处理运动规划解决方案的强化学习工具。这一结果将使混合可重构机器人能够通过复杂的估计方案在不确定性估计下部署，例如依赖于环境中特征分布的视觉惯性里程计。这项工作对于减少在工业混乱的危险密闭空间内工作时发生的与工作有关的事故数量非常重要。这一成果将使机器人在关键应用中的操作成为可能，包括发电厂检查、建筑、采矿和许多其他领域，减少将人类置于危险工作环境中的可能性。利用艾伯塔省石油和天然气行业的丰富经验，该项目将开发有效的机器人工具，这些工具可以到位并在需要时部署。结果将使在危险的密闭空间中的工作更安全、更有效，并有助于降低与进入密闭空间相关的高昂成本。

According to the International Labor Organization, about 2.3 million work-related deaths occur annually worldwide. Specifically, incidents taking place when working within confined spaces contribute to thousands of deaths and disabling injuries every year. For example, Alberta's oil and gas industry reports close to 3,000 people being disabled due to confined spaces work-related accidents annually. Autonomous robots have the potential to be used as effective tools for operations in complex confined hazardous environments - both structured and unstructured. Their motion capabilities to maneuver and perform dangerous work in environments such as those encountered inside collapsed buildings, pressure vessels, silos, storage tanks, underground cable galleries, aircraft wings, manholes, pump stations, among many others are tremendous. However, major challenges related to autonomy, locomotion, and adaptability remain before robots can be effectively deployed inside such spaces where safety-critical decisions must be made despite uncertainty in the estimation, execution, and the environment. This project will develop novel hybrid mobile robot configurations capable to access spaces of diverse geometrical features and hazardous levels combined with reinforcement learning tools for whole-body tractable motion-planning solutions. The results will enable to deploy hybrid reconfigurable 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 number of work-related incidents when working inside industrial chaotic hazardous confined spaces. The results will make possible the operation of robots in critical applications including power plant inspection, construction, mining, and many others reducing placing humans in hazardous-work situations. Using the wealth of experience in Alberta's oil and gas industry, the project will develop effective robotic tools that can be in position and deployed when needed. The results will render work in hazardous confined spaces safer, more effective, and help reduce the high cost associated with confined space entry.