Aerospace control reinforcement learning course
Online course of 300 hours in reinforcement learning for aerospace control, structured in 6 modules covering reinforcement learning fundamentals, algorithms, aerospace training environments, control applications and safety, robustness and verification, with an assigned technical tutor.
Take the controls.
Run a real engineering trade-off from this field. Move the sliders, watch the model react in real time and beat the challenge.
What you will be able to do
Understand the mathematical foundations of reinforcement learning.
Apply modern algorithms such as PPO, SAC and TD3.
Build aerospace training environments.
Design reward functions for control tasks.
Compare learned policies with classical controllers.
Evaluate the safety and robustness of policies.
Develop a reinforcement learning-based controller.
Who it's for
- Control engineers applying AI to flight and spacecraft control.
- Reinforcement learning researchers looking for aerospace applications.
- GNC engineers.
- Autonomy engineers working on drones, UAS and spacecraft.
- Machine learning engineers moving into aerospace.
- Simulation developers building training environments.
- Engineering or science graduates aiming to work in the drone and UAS sector.
What you'll work with
- EASA Artificial Intelligence Concept Paper Issue 2 (guidance for machine learning applications in aviation)
- Regulation (EU) 2024/1689 (EU Artificial Intelligence Act)
- RTCA DO-178C (software considerations in airborne systems)
- ISO/IEC 22989:2022 (AI concepts and terminology)
- ISO/IEC 23894:2023 (AI risk management)
- ECSS-E-ST-60-30C (satellite attitude and orbit control systems)
- Gymnasium
- Stable-Baselines3
- JSBSim
- PyTorch
- MATLAB Reinforcement Learning Toolbox
6 modules
1.1Markov decision processes1.2Reward, policy and value function1.3Bellman equations1.4Dynamic programming1.5Exploration and exploitation1.6Relationship with optimal control1.7Applications in flight and space1.8Limitations in critical systems1.9Gymnasium environments1.10Lab: control of an inverted pendulum
2.1Q-learning and SARSA2.2Deep Q-Networks2.3Policy gradient and REINFORCE2.4Actor-critic and A2C2.5PPO2.6DDPG and TD32.7SAC and maximum entropy2.8Model-based learning2.9Hyperparameter tuning2.10Practical exercise: comparing PPO and SAC
3.16DOF modelling for RL3.2Environments with JSBSim3.3Quadcopter simulation3.4Satellite attitude dynamics3.5Reward design3.6Observation and action spaces3.7Constraints and episode termination3.8Training parallelisation3.9Environment randomisation3.10Workshop: a Gymnasium environment on JSBSim
4.1Quadcopter stabilisation4.2Trajectory tracking4.3Satellite attitude control4.4Landing of reusable rockets4.5Orbital rendezvous and docking4.6Fixed-wing flight control4.7Obstacle avoidance4.8Multi-agent swarm control4.9Comparison with PID and MPC4.10Case study: powered landing of a launcher
5.1Safe and constrained RL5.2Safety shields and supervisors5.3Robustness against disturbances5.4Sim2real transfer5.5Formal verification of neural networks5.6Hardware-in-the-loop testing5.7Policy explainability5.8EASA requirements for machine learning5.9Deployment on embedded hardware5.10Case study: a safety supervisor for a drone
6.1Definition of the control problem6.2Model and environment6.3Reward design6.4Agent training6.5Comparison with classical control6.6Robustness testing6.7Safety analysis6.8Technical documentation6.9Report and presentation6.10Final project: final review with the tutor
How you'll learn
The programme is 100% online on the SEIUM campus, with downloadable material, recorded classes and an assigned technical tutor. There are no fixed timetables: you progress at your own pace within each intake, so you can combine it with your job.
Each module combines theory, guided reading of the original standard and applied exercises. The working software is Gymnasium, Stable-Baselines3, JSBSim, PyTorch and MATLAB Reinforcement Learning Toolbox, in free, academic or evaluation versions. Exercises are based on real procedures: stabilising a quadcopter with PPO, satellite attitude control with SAC and a simulated powered landing of a launcher. Assessment is continuous by module plus the final project.
Standards covered: EASA Artificial Intelligence Concept Paper Issue 2, Regulation (EU) 2024/1689, RTCA DO-178C, ISO/IEC 22989:2022, ISO/IEC 23894:2023, ECSS-E-ST-60-30C.
Build it end to end
Your final project follows the real engineering workflow, step by step, reviewed by your tutor.
Where this takes you
Requirements
- No prior qualification is required.
- An internet connection and a computer capable of running the programme software.
- English sufficient to read technical standards.
Ready to start?
Send your application today. Our admissions team replies within 72 hours with your study plan and funding options.
Frequently asked questions
The programme totals 300 hours over 3 months. You set your own pace within each intake, and your tutor helps you plan the workload around your job.
You work with Gymnasium, Stable-Baselines3, JSBSim, PyTorch, MATLAB Reinforcement Learning Toolbox, using free, academic or evaluation versions. Your tutor indicates the free alternative wherever one exists.
A SEIUM Institute certificate, awarded on passing the continuous assessment and the final project. It is SEIUM's own qualification, not an official state degree; its recognition depends on the programme completed and the regulations of each country.
Yes. Modules passed are recognised in a higher programme in the same area (postgraduate diploma or master's). Ask about the recognition policy when you enrol.
No prior qualification is required; you need an internet connection, a computer capable of running the programme software and English sufficient to read technical standards.
Start dates are set by SEIUM's faculty for each intake. Our admissions team will confirm the next available date when you apply.
Related programmes
Sustainable Fuels Integration Course
Online course of 300 hours in sustainable fuel integration. 6 modules covering the fuel supply chain, blend quality control, aircraft and engine compatibility, emissions accounting and…
Swarm drone course (cooperation)
Online course of 300 hours in swarm drones (cooperation): 6 modules covering aerial multi-agent systems, formation modelling and control, task planning and allocation, cooperative communications and…
LOSA Incident Management Course
Online course of 300 hours in LOSA incident management. 6 modules covering the LOSA programme, the threat and error management model, flight deck observation, LOSA data…
More in Aerospace, UAM, Drones & Space
All Aerospace, UAM, Drones & Space programmes → · How to apply · Fees & financing · Scholarships