SPIN - Astroscale - Reinforcement Learning-Based Lunar Landing Simulation
SPIN
- Closing: 11:59pm, 11th Apr 2025 BST
Perks and benefits
Candidate happiness
8.43 (1692)
8.43 (1692)
Job Description
Job Description:
Lunar landing is a classic control problem where a lander must navigate and land safely on the surface while minimizing fuel consumption and ensuring stability. This project aims to develop a reinforcement learning (RL) model using OpenAI Gym's LunarLander-v2 environment to simulate and optimize lunar landings. The project will involve a literature survey on RL algorithms, implementation in Python, and performance evaluation of various approaches.
During this internship, a successful candidate will:
Develop a reinforcement learning agent capable of autonomously landing a lunar module.
Explore different RL algorithms, such as Deep Q-Networks (DQN), Proximal Policy Optimization (PPO), and Advantage Actor-Critic (A2C).
Compare the performance of these algorithms based on landing accuracy, fuel efficiency, and training time.
Implement and test the solution using OpenAI Gym's LunarLander-v2 environment.
Provide insights into the effectiveness of RL for real-world autonomous landing scenarios.
The successful applicant should have the following qualities and abilities:
Academic:
Currently enrolled in an undergraduate program in Aerospace, Space or Machine learning or other relevant discipline
Minimum requirements:
Python programming language skills
Passionate about Astroscale’s mission to make space safe, sustainability and secure.
Enrolled in undergraduate or graduate degree
Available full-time during the summer
Proficiency in Microsoft Office Suite
Good interpersonal, organisational, and written/verbal communication skills including to both technical and non-technical audiences
Demonstrated ability to collect, evaluate and synthesize multiple sources of information, e.g. articles and research papers
Analytical mindset and strong quantitative skills – ability to think critically and creatively
Ability to appreciate both technical and commercial drivers to assess feasibility.
Desirable requirements:
Relevant previous internships or project experience
Process details:
8 weeks minimum fixed term contract to be agreed with successful candidate
5 days holiday to be taken during the placement
Working at Astroscale:
We aim to deliver cutting edge technologies that will become part of routine commercial space operations by 2030.
Astroscale is the first private company with a vision to secure the safe and sustainable development of space for the benefit of future generations, and the only company dedicated to in-orbit servicing across all orbits. Founded in 2013, we are developing innovative and scalable solutions across the spectrum of in-orbit servicing missions, including End of Life services, Active Debris Removal, In-situ Space Situational Awareness and Life Extension Services, to create sustainable space systems and mitigate against the hazardous build-up of debris in space.
Astroscale is an Equal Opportunities employer and embraces a diverse workforce. All qualified applicants, including minorities, women and individuals with disabilities are encouraged to apply.
Removing bias from the hiring process
Applications closed Fri 11th Apr 2025
Removing bias from the hiring process
- Your application will be anonymously reviewed by our hiring team to ensure fairness
- You’ll need a CV/résumé, but it’ll only be considered if you score well on the anonymous review
Applications closed Fri 11th Apr 2025