Modelling and Current Control of Supercapacitor Controller for Lightweight Robots

This paper presents a modelling and control method of a Supercapacitor controller for lightweight robots. We develop an averaged state-space model that involves arbitrary number of parallel capacitor branches with their equivalent series resistance (ESR) considered, and derive a direct transfer function from duty cycle to supercapacitor current. Based on power stage transfer function, We synthesise a Type-III compensator for closed-loop current regulation. PLECS simulations on a six-branch capacitor network show that the controller (i) limits battery peak current and reduces ripple under dynamic loading, (ii) achieves fast current transitions (0.4 ms) during load steps, and (iii) performs regenerative braking energy recovery while keeping the battery current regulated. Our approach reduces tuning complexity and enhances robustness, while ensuring predictable system response.