Gerasin O. Models and means of the control systems of multipurpose mobile robots on ferromagnetic surfaces

Thesis is devoted to actual issues of development of models and different means for increasing the efficiency of multipurpose mobile robots (MMRs) control systems capable of moving a working tool on ferromagnetic surfaces (FSs). Mathematical models of MMR with caterpillar and magnetically operated mowers, which take into account the interrelation between direction and speed with the values of clamping force (CF) and the slope of the working FS, were developed for the first time. The structure of the caterpillar MMR’s interrelated control system contains cross-couplings in the course and speed control channels as well as the joint neural regulator of spatial movement, which ensures achievement of high quality control indicators, has been improved. Neuro-fuzzy mathematical models of CF observers for MMR’s clamping electromagnets are newly synthesized on the basis of neuro-fuzzy computing systems with different types of membership functions of the linguistic terms of input variables that are able to adapt to the non-ferromagnetic nature uncertainty of the working surface of and allow to pre-identify the CF value before performing another step by the robot. The method of the manipulation object slipping control in the adaptive gripper of the MMR and the MMR itself on the sloping FS has been further developed, which allows to determine the slippage direction using a resistive slip sensor, increase the speed of adaptation of gripping compressive force and the MMR’s CF to the a priori unknown mass of the object.