Maximizing Consistent Force Output for Shape Memory Alloy Artificial Muscles in Soft Robots

TL;DR

The paper tackles the challenge of achieving consistent, high-force output from shape memory alloy actuators in soft robots, where fatigue and thermal dynamics hinder real-world deployment. It introduces a hardware framework with hot-swappable SMA bundles, a fatigue-testing protocol, and a supervisory temperature controller that caps heating to maximize force over hundreds of cycles; it also models temperature dynamics with $T_{k+1}=a_1T_k+a_2u_k+a_3$ and estimates a long-life limit via exponential fits, yielding a conservative operating temperature around $T^{SET}\approx 140^\circ$C. The main contributions are an in-situ lifetime vs loading analysis, a two-layer actuating control that prevents fatigue while maximizing output, and hardware validation showing stable force and displacement across numerous cycles. The findings enable practical deployment of SMA actuators in soft robotics by providing a quantifiable, conservative operating regime and a methodology to characterize and constrain actuator behavior in-situ, with implications for larger, safer, and more capable SMA-powered soft robots.

Abstract

Soft robots have immense potential given their inherent safety and adaptability, but challenges in soft actuator forces and design constraints have limited scaling up soft robots to larger sizes. Electrothermal shape memory alloy (SMA) artificial muscles have the potential to create these large forces and high displacements, but consistently using these muscles under a well-defined model, in-situ in a soft robot, remains an open challenge. This article provides a system for maintaining the highest-possible consistent SMA forces, over long lifetimes, by combining a fatigue testing protocol with a supervisory control system for the muscles' internal temperature state. We propose a design of a soft limb with swap-able SMA muscles, and deploy the limb in a blocked-force test to quantify the relationship between the measured maximum force at different temperatures over different lifetimes. Then, by applying an invariance-based control system to maintain temperatures under our long-life limit, we demonstrate consistent high forces in a practical task over hundreds of cycles. The method we developed allows for practical implementation of SMAs in soft robots through characterizing and controlling their behavior in-situ, and provides a method to impose limits that maximize their consistent, repeatable behavior.

Figures (8)

• Figure 1: Example comparison of force output, in a practical task lifting a 50g mass, after 150 cycles. The SMA whose states (temperature) are controlled at proposed long-life limit generates more force and displacement than another actuated past the fatigue limit.
• Figure 2: Limb fabrication process with the incorporation of the module (a). The SMA is crimped to a wire which is bonded to the module along with the thermocouples using Sil-Poxy. The created bundle (b) is threaded through the limb with the ground/pink wire (c). Once the bundle is inserted all the way through the limb (d), the SMAs and ground wire are crimped at the tip to secure the bundle in the limb (e).
• Figure 3: Testing and limb setup for recording force output. The bundle, composed of the module (1), thermocouples (2), and SMA wires (3), is placed in the center configuration of the limb then placed into a 3D printed bracket (a) that limits the movement of the limb towards the force plate. The limb is then fitted with a foot at its end (a) to localize force output and placed into contact with the force plate (b).
• Figure 4: Logical architecture for the proposed system. The SMA temperature and force output are collected and utilized in the input generator and the supervisory controller. In each experiment trial, one of the two input generators is selected to determine when to start/stop heating in every heating cycle. The generators' PWM input is then limited by the temperature supervisor to make sure the SMA stably reaches the desired temperature.
• Figure 5: Example results from controller $C_1$ and controller $C_2$ with temperature maximum of $150^\circ$C. Plot (1) shows reaching the limit temperature (1a), holding the limit temperature for 20 seconds (1b), and then cooling to $35^\circ$C (1c) as set by $C_1$. Plot (2) shows the fixed 45 seconds of heating (2a) and 65 seconds of cooling (2b) as set by $C_2$.