Machine Learning for Robotics and Control

机器人和控制的机器学习

• 批准号: CRC-2017-00284
• 负责人: Schoellig, Angela
• 金额: $ 8.74万
• 依托单位: University of Toronto
• 依托单位国家: 加拿大
• 项目类别: Canada Research Chairs
• 财政年份: 2020
• 资助国家: 加拿大
• 起止时间: 2020-01-01 至 2021-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=695305
• 关键词: Machine; Learning; Robotics; Control

Robotics is the fastest growing industry in the world. In contrast to their early industrial counterparts, the next generation of robots (such as autonomous cars, drones, and assistive robots) is expected to operate alongside humans in unpredictable and changing environments. This requires new robot algorithms that enable robots to adapt to the environment during operation. The proposed research will combine key technologies from control theory, machine learning and optimization to enable safe, high-performance robot control for next-generation robotics applications. The resultant algorithms will allow Canadian companies to build service/personal robots with enhanced capabilities in less time and at lower cost.

机器人是世界上增长最快的行业。与早期的工业机器人相比，下一代机器人（如自动汽车、无人机和辅助机器人）预计将在不可预测和不断变化的环境中与人类一起工作。这需要新的机器人算法，使机器人能够在操作过程中适应环境。拟议的研究将结合联合收割机的关键技术，从控制理论，机器学习和优化，使安全，高性能的机器人控制下一代机器人应用。由此产生的算法将使加拿大公司能够在更短的时间内以更低的成本建造具有增强功能的服务/个人机器人。