TL;DR
This paper presents a novel piezoelectric soft robot that achieves inchworm-like locomotion by dynamically controlling its friction with the ground through shape manipulation. By analytically modeling and experimentally validating how piezoelectric actuators can alter the robot's contact force profile, they demonstrate precise forward and backward crawling without specialized friction pads.
What Problem Does This Paper Address?
Traditional soft robots often rely on complex pneumatic systems or specialized surface features for locomotion, which can limit their miniaturization, speed, and adaptability to varied environments. While piezoelectric actuators offer advantages in terms of speed, size, and responsiveness, their application in soft robotics for complex motions like inchworm crawling has been limited. This work addresses the challenge of achieving controlled, inchworm-like motion in a thin, flexible piezoelectric robot by leveraging the inherent properties of piezoelectric actuation to dynamically tune the robot's friction with the ground, eliminating the need for external mechanisms like suction cups or dedicated friction pads.
Methods
The core of this work lies in the development of a comprehensive analytical model that predicts the shape of the piezoelectric soft robot under various voltage inputs, crucially incorporating the effects of gravity. This model is built in a "bottom-up" approach, starting with a single clamped actuator and progressing to a five-actuator system. The model accounts for:
- Piezoelectric Effect: The bending of the PZT actuators due to applied voltage.
- Distributed Load: The effect of gravity on the flexible robot body.
- Discrete Forces: The interaction with the ground, specifically the contact force.
A key innovation is the self-consistent approach to determine which parts of the robot lift off the ground, enabling the prediction of the "suspended length" ($L_{sus}$). This model is then used to derive the contact force distribution and, consequently, the friction asymmetry between the robot's ends. The experimental validation involves fabricating a robot from commercially available piezoelectric actuators and steel substrate, and then systematically comparing the predicted shapes, contact forces, and inchworm motion with experimental measurements.
Datasets
Not discussed in the paper. The paper relies on analytical modeling and experimental measurements using a custom-fabricated robot.
Key Insights and Ideas
- Friction Asymmetry via Shape: The most significant insight is that by strategically lifting one end of the robot off the ground using piezoelectric actuators, the contact force distribution can be altered. This "seesaw effect" creates a friction asymmetry, allowing one end to "stick" while the other "slips," mimicking the fundamental principle of inchworm locomotion.
- Gravity's Crucial Role: The authors emphasize that gravity's effect on the robot's shape is significant and critical for accurate modeling, especially given the robot's inherent elasticity. Ignoring gravity leads to inaccurate predictions of contact points and suspended lengths.
- Analytical Model Predicts Complex Behavior: The developed analytical model, incorporating piezoelectricity and gravity, accurately predicts the robot's shape, contact force distribution, and ultimately, its inchworm motion, without requiring any adjustable fitting parameters. This provides a powerful tool for designing and controlling such robots.
- Scalability and Design Optimization: The paper demonstrates how the model can be used to understand scaling effects (length and thickness) and to optimize actuator design (e.g., substrate thickness for maximum bending).
Key Results
- Accurate Shape Prediction: The analytical model shows excellent agreement with experimental measurements of the robot's shape under various voltage conditions, even with gravity considered.
- Contact Force and Friction Asymmetry: The model successfully predicts the shift in contact force and the resulting friction asymmetry when one end of the robot is lifted. Experiments validate that lifting actuator #1 increases its friction significantly.
- Inchworm Motion Achieved: The robot successfully demonstrates controlled forward and backward inchworm motion, with experimental results closely matching model predictions.
- Forward motion: ~0.78 mm/cycle.
- Backward motion: ~0.66 mm/cycle.
- Mass and Gravity Effects Quantified: Increasing the robot's mass by a factor of four is shown to reduce vertical displacement by approximately 2x, and gravity is shown to reduce lateral movement per cycle by ~22%.
- Discrete Mass for Enhanced Friction: Adding discrete masses to the robot's ends can boost the friction asymmetry, increasing the difference in contact force ratio from ~30% to ~70%.
Limitations and Future Work
- Quasi-Static Assumption: The current model and analysis are primarily quasi-static. Dynamic effects become significant at higher frequencies (~1 Hz), leading to more complex motion and potential speed increases.
- 2D Model: The current model is 2D and does not account for torsional distortions (twisting) that could occur in a narrow robot. While experiments suggest this is a minor effect for their current design, it could be relevant for different aspect ratios.
- Material Properties: The model relies on material parameters from datasheets, and experimental uncertainties (e.g., epoxy thickness) are acknowledged.
- Future Work: The authors suggest several avenues for future research:
- Higher Speeds: Exploring routes to higher speeds through stronger piezoelectric materials, exploiting dynamic effects (resonance), and investigating phenomena like jumping.
- Surface Variability and Inclines: Understanding how the robot's shape-controlled friction can handle surface variations and inclines.
- Integration: Developing untethered operation with onboard batteries, control electronics, and wireless communication.
- Complex Environments: Adapting the robot for operation in confined spaces or for tasks like swimming.
- Dynamic Modeling: Developing models to understand and exploit high-frequency dynamic effects for faster locomotion.
Summary updated