MPC Rendezvous and Contact Simulator

An entry-level MPC-based microgravity simulator

This repository provides the simulation and control framework developed during the Master's thesis "Model Predictive Control-based Rendezvous and Contact of Autonomous Space Vehicles", by Rodrigo Silva. It models various autonomous space vehicle scenarios in a planar or 3D microgravity environment, using Model Predictive Control (MPC) with CasADi.

Developed as part of a joint research effort between KTH Royal Institute of Technology (Sweden) and Instituto Superior Técnico, University of Lisbon (Portugal).

---

Features

• 2D and 3D simulation of spacecraft rendezvous and docking
• Entry angle constraints and obstacle avoidance
• Configurable test scenarios
• MPC controller using CasADi and IPOPT
• Plotting and animation tools (optional)

---

Repository Structure

MPC_simulator/
├── run_simulator.py           # Main entry point to run scenarios
├── requirements.txt           # Required Python packages
├── scenarios/                 # All scenario definitions
├── src/                       # Source code
│   ├── controllers/           # MPC controller definitions
│   ├── dynamics/              # System dynamics (2D and 3D)
│   └── utils/                 # Quaternion tools and helpers
└── results/                   # Optional folder to store simulation outputs

---

Installation

Make sure you have Python 3.12.6 installed.

Install the required packages using:

pip install -r requirements.txt

---

Running a Scenario

The scenarios can be launched in two different ways: either directly inside the scenario script (example 1) or by launching run_simulator.py (example 2). The interface will soon support command-line arguments, but for now, modify the script directly to choose:

example 1:

python scenarios/obs_avoidance.py

example 2:

python run_simulator.py mpc_rhapsody

• The simulation dimensionality (2D or 3D)
• The desired scenario name

Example scenario files:

• entry_angle_v2
• obs_avoidance
• entry_plus_obs
• dock_alignement
• fugue
• multistage
• ros_scenario

To view animations and plots, make sure to set the show = True flag inside the scenario script. To save, the same logic goes for the save = True flag inside the scenario script.

---

Reproducing Thesis Results

A mapping between each thesis figure and its corresponding scenario configuration will be provided in results/reproducibility.md. This will allow others to replicate the experimental results, both from simulations and laboratory data.

---

License

This project is released under the MIT License, a permissive license that allows reuse with attribution. It is suitable for academic and research purposes. If you're unsure or have a specific use case, feel free to reach out.

---

Author

Rodrigo Silva
This simulator was developed as part of the Master's thesis:
Model Predictive Control-based Rendezvous and Contact of Autonomous Space Vehicles
KTH Royal Institute of Technology & Instituto Superior Técnico (University of Lisbon)