| Proj No. | A1107-251 |
| Title | Large-language model based human interactive global route planning using Open Street Map |
| Summary | LLM-based global planning with OpenStreetMap (OSM) is transformative for urban autonomous navigation, such as in delivery robots, by enabling intuitive interpretation of human instructions (e.g., “go to the cafe near the park and buy me a drink”) into precise geospatial coordinates, leveraging OSM’s scalable 2D framework to generate optimal routes. However, challenges include LLMs’ difficulty in resolving ambiguous language into accurate coordinates amidst OSM’s local inaccuracies, compounded by urban dynamics like temporary obstacles or traffic. This research will harness LLMs to reason about navigation instructions, converting them into location coordinates, and utilize OSM to produce efficient navigation paths, aiming to improve accuracy and adaptability through dynamic route optimization in complex urban environments. Candidates should possess familiarity with, or an interest in, ROS development. To ensure the success of the dissertation, bi-weekly meetings will be held. *Ensuring sufficient participation time is necessary* |
| Supervisor | Prof Xie Lihua (Loc:S2 > S2 B2C > S2 B2C 94, Ext: +65 67904524) |
| Co-Supervisor | - |
| RI Co-Supervisor | - |
| Lab | Delta-NTU Copr Lab (Loc: S2-B3c-13 (Tel: 6592 2661)) |
| Single/Group: | Single |
| Area: | Intelligent Systems and Control Engineering |
| ISP/RI/SMP/SCP?: |