CAREER: Intelligent Planning and Control of Robotic Systems in a Perceptive Frame

职业：感知框架中机器人系统的智能规划和控制

• 批准号: 9703149
• 负责人: Ning Xi
• 金额: --
• 依托单位: Washington University
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 1997
• 资助国家: 美国
• 起止时间: 1997-09-01 至 1998-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=9703149&HistoricalAwards=false
• 关键词: CAREER; Intelligent; Planning; Control; Robotic

The objective of this research is to develop and experimentally test new methodologies for task scheduling, action planning and control of robotic and manufacturing systems. The new methodologies will be a combination of heuristic and analytical approaches and based on a direct algorithmic connection between simple on-line perception of not fully predictable geometric, dynamic and time events, and the desired actions of robotic or manufacturing systems. They will provide a robust and reliable method for handling local geometric and dynamic uncertainties as well as coping with unexpected events, so that the local disturbances do not become global ones causing the need for re- planning the operation. This development will help to eliminate or lessen some of the down-time of automated robotic or manufacturing systems caused by inflexible scheduling, planning, sensing and control methods. It will also help to substantially reduce the fixturing or set-up requirements of robotic or manufacturing systems. As an integral part of the research, the education activities include developing a Robotics Collaboratory Environment (RCE) which will provide a new instruction method and laboratory environment for teaching robotics, manufacturing automation and intelligent systems. As a result, more students from groups that have been traditionally underrepresented in science and engineering can be involved, and more opportunities for undergraduate students to incorporate research activities into their course work will be provided.

本研究的目的是开发和实验测试新的方法，任务调度，行动规划和控制 的 机器人和制造系统。 的 新的方法将是启发式和分析性方法的结合，并基于对不完全可预测的几何、动态和时间事件的简单在线感知与机器人或制造系统的期望动作之间的直接算法联系。 它们将提供一种稳健可靠的方法，用于处理局部几何和动态不确定性以及应对突发事件，以便局部干扰不会变成全局干扰，从而导致需要重新规划操作。 这将有助于消除或 减少自动化机器人或制造系统因不灵活的调度、规划、传感和控制方法而导致的停机时间。 它还将有助于大幅降低机器人或制造系统的夹具或设置要求。 作为研究的一个组成部分，教育活动包括开发机器人合作实验室环境（RCE） 为机器人学、制造自动化和智能系统的教学提供了新的教学方法和实验环境。因此，更多的学生从群体一直在科学和工程传统上代表性不足，可以参与，并为本科生将研究活动纳入他们的课程工作将提供更多的机会。