Tensegrity and Soft Robotics

Our lab builds structurally compliant robots for applications including exosuits, exoskeletons, manipulators, prosthetics, and rovers.

Tensegrity is a design principle that features compression elements suspended within a network of tension elements. This fusion results in a hybrid soft-rigid structure that easily complies with external stresses by distributing loads throughout the entire tension network.

This paradigm is particularly useful for developing robots that mimic musculoskeletal kinematics and dynamics. We have chosen to apply these ideas to our own biomechanically-oriented projects.

(Click to see the movie) 


Tensegrity Joints  [C57, C59, C60]

Collaborators: Vytas Sunspiral and Adrian Agogino (NASA Ames)
Researchers: Jonathan Bruce, Steven Lessard, Leya Breanna Baltaxe-Admony, Shaurya Chopra, Ash Robbins

Tensegrity Shoulder JointTensegrity Elbow Joint

Most traditional robotic mechanisms feature inelastic joints that are unable to robustly handle large deformations and off-axis moments. As a result, the applied loads are transferred rigidly throughout the entire structure. The disadvantage of this approach is that the exerted leverage is magnified at each subsequent joint possibly damaging the mechanism. Our lab has developed two lightweight, elastic, bio-inspired tensegrity robotic arms adapted from prior static models which mitigate this danger while improving their mechanism's functionality. Our solutions feature modular tensegrity structures that function similarly to the human elbow and the human shoulder when connected. Like their biological counterparts, the proposed robotic joints are flexible and comply with unanticipated forces. Both proposed structures have multiple passive degrees of freedom and four active degrees of freedom (two from the shoulder and two from the elbow). The structural advantages demonstrated by the joints in these manipulators illustrate a solution to the fundamental issue of elegantly handling off-axis compliance. Additionally, this initial experiment illustrates that moving tensegrity arms must be designed with large reachable and dexterous workspaces in mind, a change from prior tensegrity arms which were only static. These initial experiments serve as an exploration into the design space of active tensegrity structures, particularly those inspired by biological joints and limbs.

Click to see the videos: (1) (2) (3)  


Tensegrity Quadruped [C61]

Colaborators: Vytas Sunspiral and Adrian Agogino (NASA Ames)
Researchers: Dawn Hustig-Schultz

MountainGoat is a tensegrity quadruped actuated by means of a two-level control system, with impedance controllers driving the lower level reflexes of individual muscles and central pattern generators (CPGs) coordinating the higher level phase-coupled oscillations of groups of neighboring muscles. A neural network feedback to this dual-level control system. In simulations of locomotion, both CPGs and Neural Network learn the necessary parameters for robust control through machine learning using the Monte Carlo method followed by genetic evolution.

Although preliminary work has focused on simulations of primarily spine-driven locomotion on flat ground, these initial results hint at good potential for improved locomotion when CPG control is extended to the legs. The potential for rugged terrain traversal can also be seen in the robust passive terrain interaction of the robot in balancing on blocks and hills. This interaction is due to how it naturally adapts to complex footing by utilizing the multiple degree-of-freedom compliance of its tensegrity spine.

Preliminary results have shown how changing the morphology of the spine and legs has led to an increase in spinal torsion and ground reaction force, which has increased the distance traveled by the robot on flat ground. Current work is focusing on extending control to legs and improving leg lift, for traversal over uneven terrain.



Bio-Inspired Spider Robot:

Researcher: Samira Zare

Designing a silicon-based spider robot in order to access to the places and environments that rigid robots are unable to access. Flexibility and lightweight properties of this robot make it perfect for rescues missions and exploring space. Several designs, simulations, and stress analyses have been made in order to make the spider leg that has optimal movements.

spider.png2016-09-01 (12).png


Tensegrity Heat Shield:

Researchers: Steve Lessard, Shaurya Chopra

We simulated heatshields made from tensegrity structutures, capable of adjusting its shape to be aerodynamically optimal at all times. Reducing air resistance and friction reduces the heat generated during atmospheric entry, protecting space shuttles. Simulations were done in NTRT.


SUPERball Escape Algorithm:

Researchers: Jonathan Bruce, Steven Lessard

We developed control strategies for SUPERball, a tensegrity rover, to escape a crater via a multi-level Monte Carlo genetic algorithm. This could be useful for missions that require SUPERball to explore crevassed terrains, such as those on Saturn's moon Titan.