Collaborative Research: Kinetic Energy Storage and Release for High-Power Robotic Jumps and Maneuvers

This project enables powerful new locomotive capabilities for centimeter/kilogram scale robots using flywheel-based jumping. Jumping can allow robots to move through rugged terrain, reach high vantage points, cross gaps, and perform dynamic maneuvers that are difficult for wheeled or legged systems, without the continuous baseline power requirements of active flight. State-of-the-art systems capable of jumps exceeding tens of body lengths in height typically cycle through a relatively slow phase where potential energy is stored in a deforming elastic structure acting as a spring, followed by a relatively fast phase in which the deformed structure returns to its original shape and the stored elastic energy is released as translational velocity. The robots studied in this project instead store energy as rotational kinetic energy in a spinning flywheel, then convert that rotational energy to translational velocity through a mechanical transmission. This scheme potentially bypasses multiple limitations of elastic energy storage, including the limited strain energy density storage capacity of available materials, the fatigue caused by repeated high-strain loading, and the loss of performance due to non-deforming structural elements and components such as motors, batteries, electronics, and sensors and other payloads. The activities of this project will explore the new research questions raised by flywheel-based jumping, to realize the promised potential of this innovative approach. Applications of flywheel-based jumping robots include search and rescue, structural inspection, environmental monitoring, and exploration of rugged and remote environments. These important new robot capabilities, augmented by accessible hands-on demonstrations and classroom activities, will benefit the American public, develop the robotics workforce, and advance the nation's scientific leadership, safety and security, and industrial competitiveness. The technical goal of this project is to establish the principles, models, and experimental methods needed to create and control robots that use kinetic energy storage for high-power jumping and maneuvering. The research has three integrated thrusts. First, the project will develop and experimentally validate models that describe the specific energy of both the flywheel and the full robotic system in terms of flywheel shape, size, material, motor power, surrounding fluid, and embedded payloads. Second, the project will create and test nonlinear mechanical transmissions that convert flywheel rotation into linear motion with high efficiency while managing force and torque profiles during takeoff. Third, the project will develop planning and feedback control methods that exploit the unique properties of flywheel-powered robots. These methods will include energy-aware path planning for rough terrain and orientation control during takeoff and flight using inertial and gyroscopic effects. Together, these efforts will provide a design framework for mobile robots with high intermittent power demands and will enable a new class of electrically powered flywheel-based jumping robots. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.