Simulation of Active Soft Nets for Capture of Space Debris

TL;DR

The paper presents a MuJoCo‑based simulator for active soft nets aimed at autonomous space debris capture, integrating net dynamics, net self‑contact, debris contact, and orbital mechanics with a four‑satellite thruster control system. It evaluates three net mechanical models (Inextensible, Shell, Saint‑Venant) and two control schemes (PID and Sliding Mode Control) within a finite‑state framework of orienting, approaching, and capturing Envisat, using Clohessy–Wiltshire dynamics to couple orbital motion. Key findings show that more compliant nets yield higher capture success and that Sliding Mode Control provides robust performance and larger contact areas, achieving up to 100% capture in tested cases, albeit with longer approach times relative to PID. The work demonstrates real‑time capable simulation, feasible fuel margins, and the potential for optimizing net design and control for practical debris removal, while calling for broader testing across debris geometries and mission scenarios.

Abstract

In this work, we propose a simulator, based on the open-source physics engine MuJoCo, for the design and control of soft robotic nets for the autonomous removal of space debris. The proposed simulator includes net dynamics, contact between the net and the debris, self-contact of the net, orbital mechanics, and a controller that can actuate thrusters on the four satellites at the corners of the net. It showcases the case of capturing Envisat, a large ESA satellite that remains in orbit as space debris following the end of its mission. This work investigates different mechanical models, which can be used to simulate the net dynamics, simulating various degrees of compliance, and different control strategies to achieve the capture of the debris, depending on the relative position of the net and the target. Unlike previous works on this topic, we do not assume that the net has been previously ballistically thrown toward the target, and we start from a relatively static configuration. The results show that a more compliant net achieves higher performance when attempting the capture of Envisat. Moreover, when paired with a sliding mode controller, soft nets are able to achieve successful capture in 100% of the tested cases, whilst also showcasing a higher effective area at contact and a higher number of contact points between net and Envisat.

Figures (7)

• Figure 1: Overview of the simulation tool's workflow. After an initial scene initialization, both the controller and the orbital dynamics modules update the force applied to the massive bodies in the simulation between every step.
• Figure 2: Schematics of the (A) finite-state machine representing the high-level control for the capture of space debris and of the three implemented states: (B) the orienting phase, (C) the approaching phase, and (D) the capture phase.
• Figure 3: The three phases of the capture of Envisat as a function of the net mechanical model and the controller used: (A) Inextensible edges and PID, (B) Shell and PID, (C) Saint-Venant solid and PID, (D) Inextensible edges and SMC, (E), Shell and SMC (F) Saint-Venant solid and SMC. The starting position of all cases is the same, and the starting relative velocity between the net and Envisat is 0.
• Figure 4: Trajectories of the four satellites with respect to the target: (A) isometric view, projection on the (B) $xy$ plane and (C) $xz$ plane, and (D) $z$ position as a function of time. Data about the thrusters include: (E) mass of the satellites at the corners of the net and (F) thrust generated by each satellite as a function of time.
• Figure 5: Time needed to achieve capture as a function of the starting positions of the for each control modality and the model used to characterize the net: (A) Inextensible edges and PID, (B) Shell and PID, (C) Saint-Venant solid and PID, (D) Inextensible edges and SMC, (E), Shell and SMC (F) Saint-Venant solid and SMC. The size of the marker is proportional to the effective area of the net upon contact with the target. The effective area is computed by the area between the projections of the four corners onto a plane perpendicular to the vector between the center of the net and the target.