Master Project on Learning-Based Robot Trajectory Prediction
Predicting robot trajectories is critical for safe navigation, swarm coordination, and autonomous decision-making in dynamic environments. This project focuses on developing and evaluating advanced deep learning models for trajectory prediction using sequential data from mobile robots. We will explore recurrent and attention-based architectures such as RNNs, LSTMs, Transformers, and state-of-the-art models like Mamba to build robust and accurate trajectory forecasting systems.
Your profile and skills:
-
Proficiency in Python and deep learning frameworks (PyTorch)
-
Strong interest in time-series modeling, sequential data analysis, and physics-based motion dynamics
-
Familiarity with robotics, motion planning, or control is a plus
-
Ability to work analytically and approach problems in a structured, research-oriented manner
-
Independent working style with a passion for team collaboration
-
Excellent communication skills in English
Application:
Please send your CV, transcript of records, and a short motivation letter describing your interest in robot trajectory prediction and sequential deep learning models to prajol.shrestha@fau.de with the subject line: “[Learning-Based Robot Trajectory Prediction Project Application 2025]”. Applications without these documents will not be considered.