cs.ROMar 6, 2026

Underactuated multimodal jumping robot for extraterrestrial exploration

Authors: Neil R. Wagner, Justin K. Yim

Organizations: Department of Mechanical Science and Engineering, University of Illinois Urbana-Champaign

Read PDF
Ask AI
Cite

Copy citation

DOINot available for this paper

Abstract

We present a rolling and jumping underactuated monopedal robot designed to explore multimodal locomotion on low-gravity bodies. It uses only two reaction wheels to control its spatial orientation with two controllers: a balancing controller which can aim the robot's jump direction on the ground, and an aerial reorientation controller which can aim the robot's leg for landing after flight. We demonstrate rolling, targeted jumping and landing, and self-righting using only three actuators total, keeping system size to 0.33m and 1.25kg. Simple switching between locomotion modes enables the system to deal with differing landscapes and environmental conditions.

Figures & tables

Fig. 1: Underactuated robot with one leg and two reaction wheels balancing on a table
Fig. 2: Robot platform from above. The 15-degree cant angle, θ\theta , of the wheels allows for balancing and aerial reorientation while maintaining differential drive features.
Fig. 3: Isometric view of robot platform displayed in CAD
Fig. 4: Rack and pinion design for jumping locomotion
Fig. 5: Robot rolls on the ground in a figure-eight path, blue trajectory overlaid from camera tracking.
Fig. 6: Robot leaning and jumping, then reorientating itself in the air to point along the velocity vector. The data collected in this test is shown in Fig. 7 and Fig. 8 .
Fig. 7: Data traces from the lean jump depicted in figure 6 . Chassis orientation Euler angles, wheel velocity and torques, and leg pointing direction are displayed. The colored regions indicate the change in robot state: blue is balancing, green is jumping, and yellow is aerial.
Fig. 8: Leg position, velocity, and commanded torque from the leg motor. The first peak is the robot jumping by extending the rack and pinion. The second peak is the landing as the leg motor provides a counter torque to dampen the impact force.
Fig. 9: Test set-up for determining impulse rejection of the balance controller. Here the robot is hit by a mass at the end on a pendulum against the highest point of the robot.
Fig. 10: Euler angles and angular velocity of the robot in balance disturbance experiments. A pendulum struck the robot in the front-back direction at 22 and 32 s with impulse magnitudes of 0.013 Nms0.013~Nms and 0.009 Nms0.009~Nms respectively. The offsets from 00 in the Euler angles are due to the balance offset of the robot since the center of mass is not located exactly along the leg.
Fig. 11: Overlay of self-righting from rolling to balancing. Top: Robot rolls forward and stops suddenly to flip over. Middle: Robot extends leg to shift weight backwards and tip upright. Bottom: Robot lifts itself off the ground and begins balancing.
Fig. 12: Euler angles during self-righting shown in figure 11 . The red background denotes the system in rolling phase, and the blue background denotes the system in balancing phase. The discontinuity in pitch is due to the system passing the domain limits of ±π/2\pm\pi/2 .
Fig. 13: Robot Modes during the trial shown in figure 14 . The robot goes from rolling to balancing, then to jumping into the air, and then landing and rolling again.
Fig. 14: Overlay of combined behaviors: Rolling and self-righting into balancing mode, then leaning and jumping up. After jumping it reorients to point along the velocity vector, and finally lands softly, returning to rolling.
More in this topic

Robot Locomotion

Robot locomotion studies how robots move efficiently and reliably across varied terrains, environments, and movement modes.

See all 225 papers

Explore similar work

CardsList

  1. Dynamic Modeling and Attitude Control of a Reaction-Wheel-Based Low-Gravity Bipedal Hopper

    Mar 11, 2026Shriram Hari, M Venkata Sai Nikhil, R Prasanth KumarRobot LocomotionLegged Robotics

    Department of Mechanical and Aerospace Engineering, Indian Institute of Technology Hyderabad, Sangareddy, India

  2. Dual-Horizon Hybrid Internal Model for Low-Gravity Quadrupedal Jumping with Hardware-in-the-Loop Validation

    Mar 9, 2026Haozhe Xu, Yifei Zhao, Wenhao Feng +6Robot LocomotionQuadruped Locomotion

  3. Towards Low-Gravity Planetary Exploration using Reinforcement Learning for Walking, Jumping, and In-flight Attitude Control

    May 23, 2026Jørgen Anker Olsen, Kostas AlexisRobotic ControlSpace Robotics

    The authors are with the Autonomous Robots Lab, NTNU, O.S. Bragstads Plass 2D, 7034, Trondheim, Norway.