Near Real Time Simulation of an Independent Steering Independent Driving Mobile Robot

Abstract

In the mobile robotics field of study, the methods of mobile robot locomotion must be fundamentally studied. The omnidirectional motion is a promising method since it has the capability of moving in any direction with full control over its orientation and translation, a method which is also referred as holonomic motion. This paper presents the kinematic modelling and control strategy of a new class of omnidirectional robot, a four-wheeled independent steering and independent driving (ISID) mobile robot. In contrast with its counterpart of omnidirectional mobile robot systems, i.e: mecanum-wheeled or omni-wheeled robots, the use of rollers decreases the efficiency by nature of the orientation of the wheel, the motion vector of each wheel may push against one another to produce the desired net motion vector. In this work, the kinematic formulation of the proposed ISID mobile robot has been derived from a rigid body motion theory. From the kinematic model, the velocity vector and the orientation vector of each wheel can be obtained. The kinematic control law has been designed such that the error vector of robot’s pose decreased exponentially. In this work, the internal controllers of wheel’s actuators are also presented. A state-space model of DC motor and PID controllers were used to simulate the the performance of the steering and the driving performances of each robot wheel. So that, the proposed simulation setup allows a near real-time robot model and its control system can be simulated. Thus, it will be beneficial for robot engineers to realise the robot design. The driving and the steering motors have been assumed using the same DC motor model. The PID control parameters as well as the kinematic control gain were obtained hypothetically by trial and error. The performances of the kinematic control have been verified in the case of positioning a robot from the initial pose to the desired pose. The simulation results have demonstrated satisfactory performances of the developed kinematic control schemes. The derived control schemes guarantees an exponential decrease of the robot pose’s error vector.