| Proj No. | A1103-251 |
| Title | Active Exploration of Mobile Robots in Simulation Using Language Models |
| Summary | Recent advancements in language models have demonstrated their potential in guiding autonomous agents through high-level reasoning and decision-making. This project explores how language models can enhance active exploration strategies in mobile robots navigating simulated environments. By leveraging natural language understanding, the robot can interpret semantic cues, reason about unexplored areas, and generate informed navigation decisions. The research will involve training a reinforcement learning model guided by language-based instructions, allowing the robot to autonomously explore unknown environments more efficiently. Students will use state-of-the-art simulation platforms such as Habitat AI or Gazebo to test their approach before transitioning to real-world robots if progress permits. Necessary computational resources will be provided by the school server. The final outcome is expected to contribute to a conference or journal paper. Candidates for this project should possess the following qualities: • Knowledge of reinforcement learning and deep learning frameworks (PyTorch/TensorFlow). • Experience with natural language processing (NLP) models. • Proficiency in Python programming. • Familiarity with robotic simulation environments (e.g., Gazebo, Habitat AI). • A commitment to conducting experimental work. |
| Supervisor | Prof Xie Lihua (Loc:S2 > S2 B2C > S2 B2C 94, Ext: +65 67904524) |
| Co-Supervisor | - |
| RI Co-Supervisor | - |
| Lab | Internet of Things Laboratory (Loc: S1-B4c-14, ext: 5470/5475) |
| Single/Group: | Single |
| Area: | Intelligent Systems and Control Engineering |
| ISP/RI/SMP/SCP?: |