A Versatile Neural Network Configuration Space Planning and Control Strategy for Modular Soft Robot Arms
Zixi Chen, Qinghua Guan, Josie Hughes, Arianna Menciassi, Cesare Stefanini
AI summary
Problem
Modular soft robot arms face significant planning and control challenges due to nonlinear dynamics, hysteresis, and cumulative errors across modules, which are exacerbated by the inaccuracy of internal sensors.
Approach
The authors introduce S2C2A, which uses a bidirectional LSTM forward model in an optimization-based planner to generate configuration trajectories, paired with a biLSTM controller that translates rough internal sensor feedback into precise motor actions.
Key results
- Accurate offline position and orientation control
- Real-time obstacle avoidance and online target interaction
- Effective state tracking using only inaccurate internal encoder feedback
- Superior performance compared to previous control baselines
Why it matters
Provides a practical, sensor-efficient control solution for complex modular soft robots in real-world environments like healthcare and cluttered spaces.
Abstract
Modular soft robot arms (MSRAs) are composed of multiple modules connected in a sequence, and they can bend at different angles in various directions. This capability allows MSRAs to perform more intricate tasks than single- module robots. However, the modular structure also induces challenges in accurate planning and control. Nonlinearity and hysteresis complicate the physical model, while the modular structure and increased DOFs further lead to cumulative errors along the sequence. To address these challenges, we propose a versatile configuration space planning and control strategy for MSRAs, named S2C2A (State to Configuration to Action). Our approach formulates an optimization problem, S2C (State to Configuration planning), which integrates various loss functions and a forward model based on biLSTM to generate configuration trajectories based on target states. A configuration controller C2A (Configuration to Action control) based on biLSTM is implemented to follow the planned configuration trajectories, leveraging only inaccurate internal sensing feedback. We validate our strategy using a cable-driven MSRA, demonstrating its ability to perform diverse offline tasks such as position and orientation control and obstacle avoidance. Furthermore, our strategy endows MSRA with online interaction capability with targets and obstacles. Future work focuses on addressing MSRA challenges, such as more accurate physical models.