机器人装配接触状态识别与规划的模糊Petri网模型

针对机器人装配接触状态识别与规划问题,提出了基于模糊Petri网模型的处理方法。首先,机器人感知的不同类型传感信息经过模糊化处理可以具有统一的描述形式,通过引入加权函数和优先级因子,传感信息数量、传感信息真实程度及传感信息重要程度三者之间的矛盾得到有效解决,从而初步建立了机器人装配接触状态识别的模糊Petri网模型。在此基础上,依据装配过程对装配接触状态识别的积极意义,进一步建立了复合模糊Petri 网模型,它能根据当前装配接触状态识别的程度规划出下一步合理的动作方案,从而实现机器人装配接触状态的自动识别与规划。最后,通过平面小孔装配实例验证了所提出模型的合理性与可行性。

FUZZY REASONING USING FUZZY PETRI NETS FOR THE MONITORING AND PLANNING OF CONTACT STATES IN ROBOTIC ASSEMBLY

A fuzzy Petri net model to deal with the monitoring and planning of robotic assembly is presented. First, multiple types of sensor signals are converted to the same form of real values between 0 and 1 using fuzzy methods, and then weight factors and priorities of sensor signals are introduced to deal with the contradiction among the number, high degree of truth and important degree of input conditions. Furthermore, dynamic information of assembly is deeply studied to strengthen the ability of the model to converge. On the basis of the works above, an efficient composite reasoning mode is constructed to perform the automatic monitoring and planning of robotic assembly. Finally, a peg-in-the-hole example is given to illustrate the reasonability and feasibility of the proposed model.