RI: Robot Manipulation Under Uncertainty

RI：不确定性下的机器人操纵

• 批准号: 0712012
• 负责人: Leslie Kaelbling
• 金额: $ 43.66万
• 依托单位: Massachusetts Institute of Technology
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 2007
• 资助国家: 美国
• 起止时间: 2007-09-01 至 2011-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=0712012&HistoricalAwards=false
• 关键词: RI; Robot; Manipulation; Under; Uncertainty

The long-term goal of this project is to enable robot manipulationin the unstructured and uncertain environments of the home and the workplace.To operate robustly and effectively when there is uncertainty, a robotmust be able to reason explicitly about how uncertain it is, andpotentially choose actions that will yield sensory information thatwill reduce the uncertainty. This project addresses the problem ofreasoning about uncertainty by using a probability distribution overpossible underlying states of the world to represent its current``belief'', and by choosing actions in virtue of the effects they willhave on the robot's belief about the state of the world. Robotmanipulation problems will be formulated as partially observableMarkov decision processes (POMDPs) to generate effective strategiesfor manipulation under uncertainty.

该项目的长期目标是在房屋和工作场所的非结构化和不确定的环境中启用机器人操纵。在不确定性时，稳健而有效地操作，机器人能够明确地推理其不确定的不确定，并能够肯定地选择会产生不确定不确定的感觉信息。 该项目通过使用概率分布来代表其当前的“ belief”，并通过使用其对机器人对世界状态的信念的影响而选择行动来解决有关不确定性的问题的问题。 机器人处理问题将作为部分观察到的决策过程（POMDP）提出，以生成在不确定性下进行操作的有效策略。