Modeling and experimental evaluation of an improved amphibious robot with compact structure

This paper describes an improved three-dimensional (3D)-printed, low-cost, multi-functional, high-maneuverability, high-concealment, turtle-inspired mobile amphibious spherical robot for environmental monitoring and data collection. The major challenge in developing such a robot lies in its limited physical size and compact structure that allows for only one type of propulsion system to be used both on land and in water. This paper focuses on the optimization of the kinematic and hydrodynamic model of the amphibious spherical robot, so as to improve the control accuracy and stability of the robot. In order to optimize some kinematic and dynamic modeling parameters of the robot, such as the drag coefficient of robot, the angular velocity and swing angle of each joint, a solid model of the 3D-printed robot was built by SolidWorks. Our simulation results and theoretical calculations confirmed the validity of the virtual model and facilitated identification of key parameters in the design. The correctness of the modeling was demonstrated by the stability of consecutive crawling and underwater movements, providing a basis for driving and controlling methods for this amphibious robot, as well as guidance for the robot's gait trajectory. Combining the robot's crawling mechanism with related simulation results, an optimized prototype of the 3D-printed amphibious spherical robot was constructed. A series of crawling experiments on a common floor were performed with the improved robot prototype, which was also done using the previous robot. The results were evaluated by a novel optical positioning system, NDI Polaris. Moreover, several experiments were carried on land crawling and underwater swimming to verify the performance of the improved amphibious spherical robot. Comparison of experimental and simulation results demonstrated the improved robot had better amphibious motion performance, as well as more potentiality and applicability to the real structures. (C) 2017 Elsevier Ltd. All rights reserved.