Adaptive Locomotion: The Cylindabot Robot

Adaptive locomotion is an emerging field of robotics due to the complex interaction between the robot and its environment. Hybrid locomotion is where a robot has more than one mode of locomotion and potentially delivers the benefits of both, however, these advantages are often not quantified or applied to new scenarios. The classic approach is to design robots with a high number of degrees of freedom and a complex control system, whereas an intelligent morphology can simplify the problem and maintain capabilities. Cylindabot is designed to be a minimally actuated hybrid robot with strong terrain crossing capabilities. By limiting the number of motors, this reduces the robot's weight and means less reinforcement is needed for the physical frame or drive system. Cylindabot uses different drive directions to transform between using wheels or legs. Cylindabot is able to climb a slope of 32 degrees and a step ratio of 1.43 while only being driven by two motors. A physical prototype and simulation models show that adaptation is optimal for a range of terrain (slopes, steps, ridges and gaps). Cylindabot successfully adapts to a map environment where there are several routes to the target location. These results show that a hybrid robot can increase its terrain capabilities when changing how it moves and that this adaptation can be applied to wider environments. This is an important step to have hybrid robots being deployed to real situations.