Scientists with MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL) are developing technology that speeds up soft pneumatic actuator design. Soft actuators perform much of the lifting and moving of physical components in smart robotics, assistive wearables, and prosthetics. Current soft actuator development design technology progresses with baby steps. Robotics engineers who design soft actuators typically start with hand-built models and spend inordinate amounts of time building, testing, and tweaking the mechanisms.

The CSAIL engineers developed PneuAct, a scalable methodology that digitally fabricates soft actuators based on computationally-created designs. A designer creates a pattern of stitches and sensors based on how the actuator is supposed to move. PneuAct follows the pattern to fabricate textile in a manner similar to a manual knitting machine. A designer then attaches standard silicon tubing to the knitted fabric to connect to the actuator. The fabric includes a combination of elastic yarn and conductive yarn that detects contact. The conductive yarn gives the knitted actuator its sense of “touch.”

This CSAIL team is working with a wide range of test designs and potential applications for the knitted soft actuators. In addition to using actuators to move things, the team is also experimenting with actuators that detect resistive pressure. Team members built a pressure-sensing sleeve, for example, that can help a wearer bend, lift, or move body parts such as an arm, knee, or elbow. Other projects focus on capacitive sensors that return information on contact. For example, the group built a robot that could detect and react only to touch by human hands.

The CSAIL scientists continue to explore potential applications for machine-knitted soft sensor actuators. CSAIL PhD student Yiyue Luo is the lead author of a paper about the research. According to Luo, “Soft pneumatic actuators are intrinsically compliant and flexible, and combined with intelligent materials, have become the backbone of many robots and assistive technologies – and rapid fabrication with our design tool can hopefully increase ease and ubiquity.”