lab logo

The Soft Robotics Control Lab integrates the embodied intelligence of soft materials with artificial intelligence to make them move. We seek to answer fundamental questions about balancing these two methods.

  • Although physical contact with softer materials are intuitively more safe than rigid ones, how do we express that safety as as mathematical goal instead, and choose optimal actions?
  • When robots are designed to conform to their environments, what morphological tradeoffs with precision and stiffness would best perform a complicated task?
  • When soft materials are computationally difficult to model, how accurate do these models need to be for control?
  • If softer robots are to assist in our everyday lives, how do we built them larger and stronger without sacrificing the benefits of softness?

We address these questions by three research directions:

  1. Simplified Modeling
  2. Feedback Control for Safety and Interaction
  3. Design for Safety and Interaction

Videos on this page may take a while to load!

1. Simplified Modeling

Our group uses both physics-based methods and data-driven approximations to efficiently model complicated soft systems.

For purposes of control, we can use discrete differential geometry methods for fast simulations, and incorporate approximations for smart artificial muscles. We have shown real-time simulation of a soft robot powered by shape memory alloy (SMA) muscles with low error. And if a soft robot is moving slowly, a helpful approximation is static equilibrium. We have also used static beam bending approximations as part of control for soft robot manipulators, and optimization-based methods for flexible cable-driven robots.


2. Feedback Control and Safety

Robot safety is much simpler to express with artificial intelligence than embodied intelligence: if a control system can guarantee that a robot’s state remains within some bounds, we consider that to be safe. We have developed verifiably-safe control systems for thermally-actuated robots, keeping temperatures bounded even during human interactions.


These controllers have been implemented in a soft legged robot that can balance without overheating its actuators:


We have also shown that computational methods, such as robustness verification and model-predictive control, can balance imperfect models with improvements from feedback:

For soft robots whose motions do not need to be exact, we could verify some safety properties in open-loop. We have developed methods for trajectory generation of soft robot limbs that can safely mimic a human’s desired motion:


We have also developed planning algorithms for dynamic motions of aquatic soft robots, which can be used online:

3. Design for Control and Interaction

Implementing these control methods in hardware requires designs with sensing and actuation for the robot’s full state. Our recent work has developed soft walking robots with these capabilities:


Our work has applied these sensing approaches for proprioception of a soft robot, so that we can detect contact by inferring the robot’s internal stresses:


Lastly, we have shown that our safe supervisory control system can be deployed to solve challenging design problems in soft robotics. In particular, by maintaining our smart muscles’ states within some bounds, we can prevent functional fatigue. This in turn assists in control by maintaining a calibrated model for a long lifetime.