论文部分内容阅读
提出了一种机器人规划生成系统的结构，该系统对机器人物理世界进行了模型化，建立了对象模型知识库，在控制机器人动作序列的生成中，模仿了人的思维方式，将最基本、最简单的最先处理，采用差异减小法逐步减小当前状态与目标状态之间的差距而逼近在达到目标状态，并提出了一种能较好解决画面问题的方法，用该方法解决画面问题比较完美。此外，提出了一种将并行推理技术应用于机器人任务规划的方法，并设计了推理程序。
A robot planning and generating system structure is proposed. The system models the physical world of the robot and establishes a knowledge base of the object model. In controlling the generation of robot motion sequence, it imitates the way of human thinking, and the most basic and the most Simple first processing, the use of difference reduction method to gradually reduce the current state and the target state of the gap between the approximation to reach the target state, and put forward a better solution to the problem of the screen, using the method to solve the picture problem More perfect. In addition, a method of applying parallel reasoning technology to robot mission planning is proposed and a reasoning program is designed.