Using the artificial bee colony optimization, crow, and genetic algorithm for identifying and optimizing the dynamic parameters of a haptic device and operator’s hand

Haptic Devices (HDs) are used to simulate virtual objects for the user. The dynamics of the HD and the opertator’s hand have a great effect on the stability and performance of the HD. In this research, the dynamic identification of the user's hand and HD has been done using methods of Artificial Bee Colony Optimization (ABCO), Crow Search Algorithm (CSA), and Genetic Algorithm (GA), during experiments in two cases. In the first case, only the dynamics of the HD has been identified, then in the second case, the mentioned methods have been generalized for the simultaneous recognition of the dynamics of the user's hand and the HD. For this purpose, first, in each case, the theoretical stability boundary is obtained as a function of the available dynamic parameters. Then, by performing experiments on the KUKA Light Weight 4 robot in two cases of presence and absence of the operator’s hand, the experimental stability boundary has been obtained. Afterwards, by using ABCO, CSA, and GA in each case the desired dynamic parameters are obtained in such a way that the difference between the theoretical and experimental stability boundary is minimized. Studies show that while all three methods have been able to identify the parameters well, the GA method produces less error and has a better performance in this field.