Design and Control of a Structurally Elastic Humanoid Robot

Abstract

In upcoming decades, humanoid robots are expected to fill the current labor crisis by completing simple behaviors such as lifting and delivering heavy objects around a factory, sorting components, and placing parts onto an engineering build for an assembly line. Unlike other robots that have a fixed base, legged robots must rely on switching contact states and apply forces to the ground to navigate which poses several control and state estimation challenges. Currently, humanoid robots are manufactured with rigid metal linkages that are extremely complex and require an expert machinist leading to cost increases. Instead, this work presents the design and control of the 3D-printed humanoid robot, PANDORA. As opposed to contemporary approaches that incorporate the elastic element into the actuator mechanisms, PANDORA is designed to be compliant under load, or in other words, structurally elastic. This design approach lowers manufacturing cost and time, design complexity, and assembly time while introducing controls challenges in state estimation, joint and whole-body control. This work specifically focuses on the low-level hardware design at the firmware level, the joint control approach which must deal with the unmodeled structural elastic linkages, the corresponding state estimation challenges that are caused by the link elasticity, and the design of an orientation-based planner towards spatial objective planning for bipedal locomotion.