基于有限状态机(FSM)的多传感器信息融合在轮式运动机器人编程中的应用

Application of Multi-sensor Information Fusion Based on Finite State Machine (FSM) in the Programming of Wheeled Mobile Robot

在轮式运动型机器人研究中,传感器是实现机器人智能化的基础。单一的传感器所获得的信息有限,因此智能机器人通常配有数量较多的传感器以满足探测和数据采集的需要。但多传感器给机器人编程带来了不小的麻烦,如果利用单个传感器信息处理的编程思路,需要综合考虑的情况繁多,使程序语言编写复杂、可读性差,而且各传感器之间丧失了整体联系,造成机器人运行过程中的失控。针对这些问题,文章叙述了基于有限状态机思想的多传感器信息融合原理,并设计了一般化的编程算法,以轮式运动型机器人在长堤型赛道上躲避悬崖的运动为例,说明基于有限状态机思想的多传感器信息融合在机器人编程实践中的具体应用的可行性。

In the research of wheeled mobile robots,sensors are the basis of intelligent robot.The information obtained by a single sensor is limited,so intelligent robots are usually equipped with a large number of sensors to meet the needs of detection and data acquisition.But multi-sensor gives the robot-programming a lot of trouble,a wide variety of circumstances need to be considered by using a single sensor information processing programming ideas,so the programming language is complicated and poor readable.Moreover,the overall contact is lost among all the sensors,causing the robot to run out of control.To solve these problems,this paper describes the principle of multi-sensor information fusion based on finite state machine theory,designs a general programming algorithm.The example based on the movement of wheeled mobile robot to avoid-cliff on the long beach track,which illustrate the application of multi-sensor information fusion based on finite state machine theory in robot programming practice.