基于改进Smith预估器的显微视觉伺服

针对视觉延迟的问题,提出采用改进的Smith预估器(MSP)来改善视觉伺服系统的控制品质。在对基于位置的动态"look-and-move" 视觉伺服系统特性分析的基础上,建立了基于MSP视觉伺服系统的结构,同时建立了视觉伺服系统的分时模型。在微操作机器人平台上进行了点到点、抗干扰和微齿轮的跟踪抓取实验。结果表明,与只有PID控制器的视觉伺服系统相比,带MSP的视觉伺服系统具有更好的控制品质。克服了由于图像采集、传输和处理导致的视觉伺服延迟而严重影响控制系统的品质局限,并克服了传统方法通过减小系统增益来增强系统的鲁棒性,但同时降低了系统响应特性的缺点。

Visual servoing with modified Smith predictor for micromanipulation

A control scheme with a Modified Smith Predicator(MSP) for the position-based dynamic look-and-move visual servoing for the control of micromanipulation was proposed.The position-based dynamic look-and-move control scheme was applied to the vision servoing system with a dual-loop control scheme,in which the local controller was used as a position-controlled device,and the outer one as a vision servoing controller.The control scheme with a similar structure to the Smith predictor called modified Smith predictor was employed to eliminate the vision delay.The detailed analysis and experiments show that the vision control system with the proposed control scheme has better dynaimics performance than the vision control system with a single PID controller.The proposed control scheme resolves the problems of vision servoing's inherent time delay induced by time consuming image acquisition transfering and processing and improves the dynamic response properties and antijamming ability of the vision servoing.