This paper presents an intuitively straightforward yet comprehensive approach in developing and controlling a Mecanum-wheeled robot (MWR), with decent path tracking performance by using a simple controller as an end objective. The development starts by implementing two computer ball mice as sensors to realize a simple localization that is immune toward wheel slippage. Then, a linearization method by using open-loop step responses is carried out to linearize the actuations of the robot. Open-loop step response is handy, as it directly portrays the non-linearity of the system, thus achieving effective counteraction. Then, instead of creating a lookup table, polynomial regression is used to generate an equation in which the equation later represents an element of the linearizer. Next, a linear angle-to-gain (LA-G) method is introduced for path tracking control. The method is as easy as just linearly maps the summation of two angles-the angle between immediate and desired positions and the MWR's heading angle, into gains to control the wheels. Unlike the conventional control method which involves inverse kinematics, the LA-G method is directly a displacement-controlled approach and does not require the knowledge of parametric values, such as the robot's dimensions and wheel radius. Finally, all the methods are implemented, and the MWR experimentally demonstrates successfully tracking various paths, by merely using proportional controllers.

The addition of omnidirectional capability and modularization to conveyor systems is an exciting and trending topic in current conveyor research. The implementation of omnidirectional modular conveyors is foreseen as mandatory in the future of conveyor technologies due to their flexibility and efficiency. In this paper, an E-pattern omniwheeled cellular conveyor (EOCC) is first introduced. Camera and image processing techniques are utilized to achieve a centralized system, which is more robust than conventional decentralized systems. In order for cartons to maneuver across the EOCC, a unique method for activation of the actuators is subsequently designed. Next, the nominal characteristic of the EOCC is analyzed based on the step responses, which then inspire the proposal of four simple controllers, namely the P–P–P–P controller, tTS + P–P–P–P controller, P–P–PD–PD controller, and tTS + P–P–PD–PD controller (P for proportional, D for derivative and tTS for time-shifting), which are used to evaluate the tracking performance of the EOCC in diagonal-shaped, ∞-shaped (horizontal lemniscus), and 8-shaped (vertical lemniscus) trajectories. The results show that there is no clear winner among the controllers, with each having its own advantages and disadvantages. Nevertheless, such findings provide clearer insight into the EOCC, which is vital for future works. After all, the introduction of the EOCC system in this paper is also anticipated to elevate the benchmarks and competitiveness in the current field of modern conveyor technologies.

No abstract

No abstract