Thesis project: Reinforcement Learning (RL) for Autonomous Forklift Testing

Toyota Material Handling Manufacturing Sweden; Mjölby kommun; On-site; SEK 156,000 - 257,000; Part time

Job summary

Toyota Material Handling Manufacturing Sweden in Mjölby, Sweden, invites applications for a Master Thesis focused on reinforcement learning to automatically generate test scenarios for autonomous forklift software. You will work with an established simulation and testing environment, defining scenario parameters and guiding the learning process toward challenging operating conditions.

The project compares RL with simpler approaches, analyzes failure scenarios and test coverage, and explores

Qualifications

• M.Sc. in Computer Science or Artificial Intelligence or equivalent.
• M.Sc. in Robotics or Autonomous Systems or equivalent.
• M.Sc. in Software Engineering, Computer Engineering, or a related engineering field.

Responsibilities

• Develop and evaluate a reinforcement learning approach for generating test scenarios for autonomous forklift software.
• Work with an existing simulation and testing environment and explore how reinforcement learning can interact with it.
• Define relevant scenario parameters and investigate how the learning process can be guided toward challenging or interesting operating conditions.
• Compare reinforcement learning with simpler scenario-generation approaches, such as random testing and other search or optimization methods.
• Analyze discovered failure scenarios and evaluate test coverage, failure discovery rate, and testing efficiency.
• Investigate whether failure scenarios discovered in simulation can be reproduced and validated on a physical autonomous truck.

Skills

Machine learning

Reinforcement learning

Programming skills

Education

M.Sc. in Computer Science / AI

M.Sc. in Robotics / Autonomous Systems

M.Sc. in Software/Computer Engineering

Tools

Simulation environment

Job description

Are you interested in reinforcement learning, autonomous systems, and finding the limits of intelligent software? At Toyota Material Handling, you will investigate how reinforcement learning can be used to automatically generate challenging test scenarios for autonomous forklift software and discover failures and edge cases more efficiently.

In your Master Thesis at Toyota, you will work on:

• Develop and evaluate a reinforcement learning approach for generating test scenarios for autonomous forklift software.
• Work with an existing simulation and testing environment and explore how reinforcement learning can interact with it.
• Define relevant scenario parameters and investigate how the learning process can be guided toward challenging or interesting operating conditions.
• Compare reinforcement learning with simpler scenario-generation approaches, such as random testing and other search or optimization methods.
• Analyze discovered failure scenarios and evaluate test coverage, failure discovery rate, and testing efficiency.
• Investigate whether failure scenarios discovered in simulation can be reproduced and validated on a physical autonomous truck.

The thesis will be carried out with support from Toyota Material Handling, including guidance on the existing testing environment, available data, and autopilot system. As this is an exploratory research project, well-founded partial results are considered valuable outcomes, and the goal is not to deliver a production-ready solution.

Validation of autonomous forklift software requires testing across a large number of operating conditions. Warehouse layouts, obstacles, pallet locations, vehicle configurations, traffic, and other environmental factors create a scenario space that is difficult to cover through manually designed test cases alone.

The purpose of this thesis is to investigate whether reinforcement learning can be used to intelligently generate test scenarios. Rather than controlling the forklift, the RL agent selects the conditions under which the autonomous forklift is tested. Based on the outcome of previous tests, the agent learns which areas of the scenario space are more likely to expose failures, unexpected behavior, or challenging edge cases.

The approach will be evaluated against simpler testing strategies to determine whether reinforcement learning can improve failure discovery, test coverage, or testing efficiency. An important outcome will be an objective assessment of when RL provides value and when simpler approaches may be sufficient.

SCOPE

2 persons / 30 ECTS each

Requirements:

• M.Sc. in Computer Science or Artificial Intelligence or equivalent
• M.Sc. in Robotics or Autonomous Systems or equivalent
• M.Sc. in Software Engineering, Computer Engineering, or a related engineering field

A background in machine learning and reinforcement learning, together with good programming skills, is beneficial. Experience or interest in autonomous systems, optimization, mechanical and electrical knowledge, simulation, or software testing is considered an advantage.

Who is Toyota Material Handling?

Toyota Material Handling is a global leader in material handling, and we are making significant investments to meet the needs of the future. At our site in Mjölby, 3,000 employees work across the entire material handling value chain, from development concepts to finished vehicles. Our product range spans from manual hand trucks to autonomous vehicles and innovative energy solutions.

A Sustainable Employer

At Toyota Material Handling, we strive to create a friendly, safe and forward-thinking workplace. Our culture is built on Toyota’s core values, where respect and consideration guide us in our daily work. Our ambition is to strengthen our competitiveness by increasing diversity across the organization and embracing our differences. Through our leading environmental initiatives, ambitious climate targets and people-focused policies, we are committed to being a sustainable employer.

Start

January 2027

Instagram: ToyotaMHsweden

Linkedin: Toyota Material Handling Manufacturering Sweden AB