[Paper Review] Inclined Surface Locomotion Strategies for Spherical Tensegrity Robots

Lee-Huang Chen, Brian Cera+9 more|arXiv (Cornell University)|Aug 27, 2017

Structural Analysis and Optimization14 references3 citations

TL;DR

This paper presents a teleoperated spherical tensegrity robot (TT-4 mini) that achieves robust locomotion on steep inclines up to 24° using a novel multi-cable actuation scheme. By simultaneously actuating multiple cables, the robot achieves significantly faster speeds and improved climbing performance over single-cable actuation, marking the first successful hardware demonstration of stable inclined surface locomotion in a spherical tensegrity robot.

ABSTRACT

This paper presents a new teleoperated spherical tensegrity robot capable of performing locomotion on steep inclined surfaces. With a novel control scheme centered around the simultaneous actuation of multiple cables, the robot demonstrates robust climbing on inclined surfaces in hardware experiments and speeds significantly faster than previous spherical tensegrity models. This robot is an improvement over other iterations in the TT-series and the first tensegrity to achieve reliable locomotion on inclined surfaces of up to 24\degree. We analyze locomotion in simulation and hardware under single and multi-cable actuation, and introduce two novel multi-cable actuation policies, suited for steep incline climbing and speed, respectively. We propose compelling justifications for the increased dynamic ability of the robot and motivate development of optimization algorithms able to take advantage of the robot's increased control authority.

Motivation & Objective

• To develop a spherical tensegrity robot capable of reliable locomotion on steep inclined surfaces, a critical need for planetary exploration.
• To overcome the limitations of single-cable actuation, which restricts incline climbing to ≤13° and limits speed.
• To design and validate multi-cable actuation policies that enhance both climbing ability and locomotion speed.
• To establish a hardware benchmark for future optimization of tensegrity robot control policies.
• To motivate the use of learning algorithms for autonomous optimization of multi-cable gaits and robot topologies.

Proposed method

• Designed a six-bar spherical tensegrity robot using a modular elastic lattice prototyping platform for rapid assembly.
• Implemented a novel two-cable actuation scheme with alternating and simultaneous control policies to enhance dynamic control authority.
• Developed a simulation framework to analyze locomotion under single-cable and multi-cable actuation, focusing on incline performance and speed.
• Conducted hardware experiments on 10° and 24° inclines to validate simulation results and benchmark performance.
• Measured the robot’s coefficient of friction to estimate theoretical maximum incline (26°), identifying friction as the primary failure limit.
• Used teleoperation to test and compare actuation policies, with performance evaluated via average velocity and incline traversal success.

Experimental results

Research questions

• RQ1 Can a spherical tensegrity robot achieve stable locomotion on inclines steeper than 13° using multi-cable actuation?
• RQ2 How does simultaneous multi-cable actuation improve locomotion speed compared to single-cable actuation?
• RQ3 What are the primary dynamic and mechanical factors enabling improved incline climbing with multi-cable control?
• RQ4 To what extent does friction limit the robot’s ability to climb steeper inclines, and can this be mitigated through design or control?
• RQ5 Can learning-based optimization algorithms further improve multi-cable actuation policies beyond human teleoperation?

Key findings

• The TT-4 mini successfully climbed a 24° incline using alternating two-cable actuation, setting a new record for spherical tensegrity robots.
• On a 10° incline, simultaneous two-cable actuation increased average velocity to 4.22 cm/s, a 115% improvement over the single-cable baseline of 1.96 cm/s.
• The robot’s maximum incline capability was limited by insufficient friction, with a theoretical maximum incline of 26° based on measured coefficient of friction.
• Failure at 24° and above was due to slipping, not loss of balance, indicating that material and surface interaction are critical design factors.
• The simultaneous actuation policy enabled faster, more fluid rolling motion, suggesting potential for further speed gains with overlapping cable contractions.
• The study demonstrates that multi-cable actuation significantly enhances control authority and dynamic performance, paving the way for AI-driven gait optimization.