Semantic Based Planning

Semantic-based planning aims to enable robots to understand and act upon high-level instructions expressed in natural language, going beyond simple geometric path planning. Current research focuses on integrating large language models with online planning algorithms, often within hierarchical frameworks that combine semantic reasoning with geometric constraints, and utilizing representations like semantic belief graphs or situational graphs to improve robustness and efficiency. This field is crucial for advancing robot autonomy in complex, unstructured environments, with applications ranging from emergency response to human-robot collaboration in shared spaces.