04 December 2023
Continuum soft robotic platforms promise to revolutionize fields where safety and robustness to uncertainty are paramount, for example enabling human-robot interaction in industry 4.0, agri-food, and medicine. Besides design and fabrication, another challenge for the development of this systems is the artiļ¬cial brains controlling the robot, especially regarding perception. Understanding the robot shape is critical for control, but accurately measuring and reconstructing the shape of a soft robot without relying on exteroceptive sensing techniques is very challenging.
Inertial measurement units (IMUs) are a feasible solution to this challenge by equipping soft robots with preconception capabilities, without altering the soft nature of the system, and without requiring the implementation of complex machine learning techniques. However, they are affected by well-known drift issues.
In this work, EMERGE partners from Delft University of Technology and collaborators propose a method to eliminate this limitation by leveraging the Piecewise Constant Curvature model assumption. The authors validate the reconstruction capabilities of the algorithm in simulation and experimentally. To this end, they also present a novel large-scale, foam-based manipulator with embedded IMU sensors. Using the filter, they bring the accuracy in IMU-based reconstruction algorithms to 93% of the soft robot's length and enable substantially longer measurements than the baseline. They also show that the proposed technique generates reliable estimations for closed-loop control of the robot's shape.
Read the paper: https://doi.org/10.1109/LRA.2023.3339063
Access EMERGE publications in the link below.
More Information
