Abstract: An autonomous jumping microrobot has been designed, and its mechanical components have been fabricated and tested. Millimeter-scale autonomous mobile microrobots have potential applications in mobile sensor networks as well as search and exploration tasks. However, mobility is difficult at this scale due to rugged surfaces, obstacles and locomotion efficiency. Jumping has been proposed as a locomotion method to overcome these challenges. The microrobot design has been divided into four components: energy storage, high work density actuators, power, and control. Like its biological inspiration, the flea, a jumping microrobot requires an energy storage mechanism to store energy and release it quickly to jump. Small leg lengths require large accelerations to reach takeoff velocities required to jump 10s of cm. Silicone micro rubber bands have been fabricated and demonstrated to store and quickly release enough energy for a 10 mg robot to jump 17 cm straight up. To stretch these micro rubber bands, electrostatic inchworm motors have been designed and fabricated to provide high forces and large displacements with low input power requirements. Three key design innovations have been used to improve the force density of these motor designs 37x over previous efforts. First, a pre-biasing actuator reduces initial electrostatic gaps below lithographic limits. Second, a toothless, friction-based clutch allows for variable stop sizes and single drive actuator motors. Third, silicon nitride has been added to reduce motor sire. Initial motor designs using these three new features have been fabricated and tested. Finally, several prototypes have been built to integrate and test the four robot components. A small-scale version of the full robot with previously fabricated solar cells and an off-the-shelf microcontroller driving a small inchworm motor has been demonstrated. Separately, an inchworm motor has been used to store energy in a micro rubber band for quick release. It is hoped that many of the design and fabrication ideas presented in this work can be used to make autonomous mobile microrobots a reality.