Bidirectional Locomotion of Soft Inchworm Crawler Using Dynamic Gaits

Liang Du, Shugen Ma, Keisuke Tokuda, Yang Tian, Longchuan LiFrontiers Robotics AI 2022 · DOI

TL;DR

This paper introduces a novel soft inchworm crawler capable of bidirectional locomotion by leveraging the robot's body deformation to dynamically switch between two distinct gaits. This approach simplifies the system architecture and actuation requirements, offering a promising new method for agile movement in soft robotics.

What Problem Does This Paper Address?

Traditional inchworm-inspired robots often require complex actuation mechanisms and sophisticated control algorithms to achieve versatile locomotion, particularly bidirectional movement. The challenge lies in effectively manipulating friction forces to generate controlled movement in both forward and backward directions without overly complicating the robot's design. Existing solutions often rely on intricate control schemes or specialized actuators. This paper aims to address this by proposing a simpler, "implicit" control strategy that utilizes the inherent dynamic properties of a soft robot's body deformation to achieve bidirectional locomotion.

Methods

The core innovation lies in the "dynamic gaits" generated by the robot's body deformation. The researchers designed a soft inchworm crawler with two integrated differential friction mechanisms. These mechanisms, combined with the robot's cyclical body deformation, allow for two distinct locomotion gaits:

  • Primitive Gait: This gait relies on the asymmetric geometric properties of the robot's body ends, leading to direction-dependent friction. It's active when the body deformation is below a critical length ($L_c$).
  • Composite Gait: This gait is activated when the body deformation exceeds $L_c$. It dynamically switches contact points between the body ends and specially designed "pad-feet." By using materials with different friction coefficients for these contact points, an asymmetric friction property is established, allowing for locomotion towards a high-friction side.

The paper proposes two control schemes based on these dynamic gaits: amplitude-based and frequency-based control. A simplified 2D mathematical model was developed to analyze the robot's dynamics, considering mass, spring stiffness, and differential friction forces. This model was then used for numerical simulations to predict locomotion performance. Finally, a prototype soft inchworm crawler was fabricated using a Dielectric Elastomer Actuator (DEA) structure and experimentally tested to validate the proposed control methods.

Datasets

Not discussed in the paper. The study relies on simulations and experimental validation with a fabricated prototype.

Key Insights and Ideas

  • Implicit Control via Body Dynamics: The most significant insight is the use of the robot's own body deformation dynamics as an "implicit controller" for locomotion. Instead of directly controlling friction, the system manipulates actuation to induce specific deformation patterns that naturally lead to desired gaits.
  • Dynamic Gait Switching: The concept of two distinct gaits (primitive and composite) that are automatically activated based on the degree of body deformation is a key enabler for bidirectional movement.
  • Simplicity in Design: The approach significantly simplifies the robot's architecture and control compared to traditional methods, relying on the inherent properties of soft materials and their deformation.
  • DEA Suitability: The paper highlights the suitability of DEA technology for this application due to its large deformation, high power density, and energy efficiency, which are crucial for generating the required body dynamics.

Key Results

  • Bidirectional Locomotion Achieved: Both simulations and experiments successfully demonstrated bidirectional locomotion. The robot could move forward or backward by adjusting actuation amplitude and frequency.
  • Control Schemes Validated: Amplitude-based and frequency-based control strategies were shown to effectively influence locomotion direction and speed.
  • Dynamic Gait Effectiveness: The composite gait, activated by exceeding the critical body length, was shown to reverse the locomotion direction compared to the primitive gait.
  • Resonance Effect: A resonance effect was observed around 4 Hz in simulations, leading to abnormally high locomotion velocities.
  • Energy Efficiency: The Cost of Transport (CoT) analysis indicated that the robot achieved relatively high energy efficiency in many operating regions, particularly at moderate actuation amplitudes and frequencies.

Limitations and Future Work

Limitations: * Model Simplification: The mathematical model used for analysis is simplified, which leads to some discrepancies between simulation results and experimental findings. * Environmental Sensitivity: The paper acknowledges that environmental factors, such as substrate surface conditions and inclines, can significantly affect locomotion performance. * Actuation Frequency Limit: The experimental setup was limited by the response time of the relays, restricting the maximum actuation frequency to around 5 Hz.

Future Work: * Advanced Dynamics Study: Further research is needed to conduct a more in-depth study of the robot's dynamics to better understand the inchworm locomotion mechanism. * Improved Robot Design: Exploring enhancements to the robot's design could lead to improved performance. * Advanced Control Schemes: Investigating more sophisticated control strategies could unlock further capabilities. * Environment Adaptation: Future work could focus on developing mechanisms for the robot to adapt to varying environmental conditions.

Summary updated

BibTeX
@article{du2022bidirectional,
  author = {Liang Du and Shugen Ma and Keisuke Tokuda and Yang Tian and Longchuan Li},
  title = {{Bidirectional Locomotion of Soft Inchworm Crawler Using Dynamic Gaits}},
  journal = {Frontiers Robotics AI},
  year = {2022},
  doi = {10.3389/frobt.2022.899850},
}
Figure