Design and Gait Optimization of an In-Pipe Robot with Bistable Inflatable Fabric Actuators

Pipe inspection robots are critical for detecting leaks or cracks, especially in environments that are harmful or inaccessible to humans. In particular, sewage pipes pose significant challenges for traditional rigid pipe robots because they can be filled with obstacles and liquid. Soft robots have been proposed to address some of the issues, but they are still limited in their ability to negotiate obstacles. In this paper, we propose a novel pipe inspection robot powered by a new class of bistable inflatable fabric actuators (BIFA). The entire robot weighs only 350 grams and can exert around 35 N of force by firmly anchoring to the pipe. It is also able to operate in pipes that are blocked by up to 34%. To understand the dynamics of the robot and simulate it for gait optimization, a reduced-order model is proposed and calibrated with characterization experiments including static loading, step response, and anchor test. A Central Pattern Generator (CPG) is also employed to parameterize the gait, enabling Bayesian Optimization in simulation to maximize the robot’s speed inside an unobstructed pipe. The optimized gait from the optimization is directly deployable on the real robot and results in a 120% speed increase over the baseline at $23 \mathrm{~mm} / \mathrm{s}$, showing the effectiveness of our model and the importance of gait selection for pipe robots.