Last Updated: 5/2/2025

For for a full publication list please see here.

2025

FLEXIV: Adaptive Locomotion via Morphological Changes in a Flexible Track Vehicle

Robotics and Automation Magazine

Sareum Kim*, Daniil Filimonov*, and Josie Hughes

Early Access

Acknowledgement : This project by Daniil Filimonov was supported by the Shakhmardan Yessenov Foundation (https://yessenovfoundation.org/en/ ), founded by Galimzhan Yessenov.

Screenshot 2025-09-05 at 10.46.58.png

Flexible and adaptive tracked robots show the capacity to nav- igate in unstructured terrain, with a passive adaption of the track or a morphology change for terrain adaptation. Extend- ing the concept to one in which the morphology can be active- ly changed has the potential to increase the capabilities of a locomoting robot while reducing the need for complex con- trollers and onboard navigation systems. This article introduc- es FLEXIV, a 232-g untethered robotic vehicle with flexible, magnet-equipped tracks that achieves adaptive locomotion across diverse geometries of ferrous terrains with appropriate transitioning of its track shape from circular to oblong. By modulating the configuration of a track loop, the robot can adjust its driving capability, such as traction and steering abili- ty, to optimize its behavior to the terrain. This deformable robot is combined with an autonomous controller that leverag- es only robot posture information through inertial measure- ment units (IMUs) for terrain estimation to autonomously adapt the robot’s configuration to the environment. This enables the robot to autonomously navigate complex terrains, including diverse slopes and steps, and offers recovery actions for extreme falls.


Robotic locomotion through active and passive morphological adaptation in extreme outdoor environments

Science Robotics

Max Polzin, Qinghua Guan, Josie Hughes Published: ****26 Feb 2025

Link to video

image.png

Robotic locomotion has advanced, but robots still lack the agility of animals in complex terrains, limiting their usefulness. Traditional robots rely on detailed perception systems, while a bioinspired approach uses morphological reconfiguration to adapt. However, such adaptability is often limited or reduces efficiency. Our study develops robots that actively reconfigure their morphology to navigate diverse environments efficiently. By combining compliance with shape-shifting abilities, a robot transitioned between flat and spherical forms, enabling autonomous driving, rolling, and swimming with minimal sensing. It successfully navigated 4.5 km across varied terrains, surpassing traditional robots in versatility, efficiency, and robustness. This innovation could enhance autonomous navigation for applications like environmental monitoring, disaster response, and space exploration.


2024

Online Hydraulic Stiffness Modulation of a Soft Robotic Fish tail for Improved Thrust and Efficiency

Soft Robotics Journal