HELP: Hierarchical Embodied Language Planner for Household Tasks
This addresses the challenge of robust planning for embodied agents in household scenarios, though it appears incremental as it builds on existing LLM capabilities with a hierarchical architecture.
The authors tackled the problem of enabling embodied agents to perform complex household tasks by planning from natural language instructions, proposing a hierarchical LLM-based planner called HELP that achieved successful real-world deployment with open-source models.
Embodied agents tasked with complex scenarios, whether in real or simulated environments, rely heavily on robust planning capabilities. When instructions are formulated in natural language, large language models (LLMs) equipped with extensive linguistic knowledge can play this role. However, to effectively exploit the ability of such models to handle linguistic ambiguity, to retrieve information from the environment, and to be based on the available skills of an agent, an appropriate architecture must be designed. We propose a Hierarchical Embodied Language Planner, called HELP, consisting of a set of LLM-based agents, each dedicated to solving a different subtask. We evaluate the proposed approach on a household task and perform real-world experiments with an embodied agent. We also focus on the use of open source LLMs with a relatively small number of parameters, to enable autonomous deployment.