Soft robotics is an area that is promising with its vast application space. One of the challenging aspects of this branch of robotics is the control of soft structures. This paper proposes a neural central pattern generator (CPG) based control architecture using an amplitude-adaptive oscillator for the movement of a low cost, pneumatically actuated soft robotic tentacle with three air chambers. The CPG is created using an SO(2) oscillator that generates half-sinusoidal outputs for pneumatic control. Through the use of an adaptation mechanism, the Dual Integral Learner (DIL), the parameters of the CPG are modulated to generate oscillatory signals of larger or smaller amplitude upon external perturbations to the system. The proposed neural control is implemented on the physical system and its validity is tested through physical restriction of the pneumatic air supply to the soft robotic tentacle.
|Conference||27th International Conference on Neural Information Processing, ICONIP 2020|
|Period||18/11/2020 → 22/11/2020|
|Series||Lecture Notes in Computer Science|
- Soft robotics
- Adaptive control