Workshop: Learning, Perception and Control in Robotics and Humans

自主和强大的智能是重要的研究领域，因为从太空探索、制造、机器人和交通到医学和生物学的应用程序过多。在航天应用中，由于深空探测的需要，自主性变得越来越重要。在制造业，安全人机和人机交互技术的发展有可能提高现有工业流程的竞争力，创造经济增长。在医学和生物学领域，使用智能系统可以改善医疗保健，将风险因素降至最低。在灾害应对领域，需要自主系统在偏远和对人类环境危险的环境中运行。鉴于自主和强大的智能在上述领域的重要性，本次研讨会旨在确定基本的科学问题，并鼓励在航空航天、医学、交通运输、制造、灾害应对、空间探索和生物学的交汇处开展新的研究方向。学习、感知和控制是自主系统在动态、不确定和偏远环境中运行所必需的基本模式。自主系统应该能够健壮地行走、导航、科学探索、快速学习新的运动技能，并将这些技能推广到看不见的条件。这个研讨会汇集了来自不同科学和工程领域的科学家，就两个问题进行集思广益，这两个问题涉及感官信息和数据的表示，以及机器人和自主系统中决策和控制机制的泛化。上述主题是在计划和控制、信息论、机器学习、神经科学和感知的交叉点上进行研究的。研讨会的重点将是基于数学概念的这些领域的数学相互依存和相互联系，这些概念包括但不限于离散几何和拓扑学。研讨会的目标是确定未来的研究方向，并确定控制理论、机器学习、感知和认知科学等学科的未决问题。

Autonomy and robust intelligence are important research areas due to the plethora of applications from space exploration, manufacturing, robotics and transportation to medicine and biology. In aerospace applications, autonomy becomes more and more important due to the need for deep space exploration. In manufacturing, the development of technologies for safe human-machine and human-robot interaction have the potential to improve the competitiveness of existing industrial processes and create economic growth. In medicine and biology, the use of intelligent systems can improve health care and minimize risk factors. In the area of disaster response, there is the need for autonomous systems to operate in remote and dangerous for the human environments. Given the importance of autonomy and robust intelligence in the aforementioned areas, this workshop aims to identify fundamental scientific questions and encourage new research directions at the confluence of aerospace, medicine, transportation, manufacturing, disaster response, space exploration, and biology.Learning, perception and control are fundamental modalities necessary for autonomous systems to operate in dynamic, uncertain and remote environments. Autonomous systems should be able to robustly walk, navigate, efficiently explore, quickly learn new motor skills and generalize these skills to unseen conditions. This workshop brings together scientists from different areas of sciences and engineering to brainstorm on two questions related to the representation of sensory information and data, and generalization of decision and control mechanisms in robotics and autonomous systems. The aforementioned topics are investigated at the intersection of planning and control, information theory, machine learning, neuroscience and perception. The emphasis of the workshop will be on the mathematical interdependencies and interconnections of these areas based on mathematical concepts that include but they are not limited to differential geometry and topology. The goal for the workshop is to determine future research directions and identify open questions across the disciplines of control theory, machine learning, perception and cognitive sciences.