Design Improvement of Wheeled Mobile Robots: Theory and Experiment

Abstract

Abstracts: In search of a solution to improve the reliability and quality of robotic systems, a variety of techniques has been developed. This study presents the empirical review of performance analysis of wheeled mobile robots using Failure Mode and Effect Analysis (FMEA) technique. FMEA is a qualitative method capable of analyzing the potential unreliability, risks and hazards during the designing and testing the system by recognition of failures with respect to critical situations. This paper addresses the design improvement and experimental validation of three prototype two-wheeled mobile robots namely preliminary, intermediate and final robots. The improvement process starts by defining three improvement measures consisting of the benchmark test radial error, the jumping height and the camera data transmission range. Each measure has its own agreed value indicating the acceptable limit. The benchmark test establishes the workability of and accuracy that is achieved with the University of Michigan Benchmark (UMBmark) odometry method. The preliminary robot is manufactured using the results obtained from robot equations and simulation studies. The robot is then tested, the results are compared with the agreed values for each test and the corrective actions are derived. The corrective actions are applied to the preliminary robot to reduce the potential risks which results in the fabrication of intermediate robot. To further ensure that the robot reaches the desired requirements, the similar process is used to achieve the final robot by using the corrective actions resulted of testing the intermediate robot. The second experiment compares the performance of mobile robots for the circular and triangular paths. The experiments illustrate that the measures possess the satisfactory improvements for final robot as compared to the preliminary one; specifically, the benchmark radial error is improved up to 45%.