
SayPlan= Grounding Large Language Models using 3D Scene Graphs for Scalable Robot Task Planning
主要的思想都在上面这个伪代码里,通过只展开部分场景图(严格层级结构),来控制输入llm的场景图大小。
A scalable approach to ground LLM-based task planners across environments spanning multiple rooms and floors
Scene Graph 通过networkx (python package)表示
Insight
- 每一次场景的展开与否,该节点是否是任务关注的节点都是由LLM决定的,这一点和我的想法一致。等于是将LLM作为一个检查器一层层遍历查找任务的兴趣点。
- Scene Graph Simulator作为任务是否可行的验证器。
SayPlan= Grounding Large Language Models using 3D Scene Graphs for Scalable Robot Task Planning