A set of system identification experiments are conducted for a modified differential drive robot. A linear model was developed by performing identification experiments to verify the model and determine unknown parameters by utilizing the motion profiles of the robot. A discrete model based on travelled distance increment was used. Nonlinear model estimates have also been generated using automated identification functions from the MATLAB's system identification toolbox. The linear model was tested through obtained data and the results were compared with the nonlinear model. It was observed that the assumption of the linearly time-invariant model allows for the state-space formulas to be implemented so that these reproductions are developed in a relatively simple structure that does not unnecessarily complicate the control system.
Cite this article as: M. Bakirci, "Data-driven system identification of a modified differential drive mobile robot through on-plane motion tests," Electrica, 23(3), 619-633, 2023.