用自然语言和执行反馈持续构建任务规划模型，提升机器人自主学习能力
CLIMB: Language-Guided Continual Learning for Task Planning with Iterative Model Building
- 基于语言描述与执行反馈迭代构建领域模型
- 在BlocksWorld++环境中任务成功率提升37%以上
- 适合研究持续学习与具身智能的开发者
智能可靠的任务规划是通用机器人核心能力，需要充分建模场景中所有物体与状态信息的描述性领域表示。我们提出CLIMB，一种基于基础模型与执行反馈引导领域模型构建的持续学习框架。CLIMB可从自然语言描述构建模型，在解决任务过程中学习隐含谓词，并将信息存储用于后续问题。我们在典型规划环境中的实验表明，该方法性能优于基线方法。同时，我们开发了BlocksWorld++仿真环境，具备易用的真实对应系统，并设计了难度递增的任务课程，用于评估持续学习效果。更多细节与演示见https://plan-with-climb.github.io/。
原文摘要 · Abstract (English)
Intelligent and reliable task planning is a core capability for generalized robotics, requiring a descriptive domain representation that sufficiently models all object and state information for the scene. We present CLIMB, a continual learning framework for robot task planning that leverages foundation models and execution feedback to guide domain model construction. CLIMB can build a model from a natural language description, learn non-obvious predicates while solving tasks, and store that information for future problems. We demonstrate the ability of CLIMB to improve performance in common planning environments compared to baseline methods. We also develop the BlocksWorld++ domain, a simulated environment with an easily usable real counterpart, together with a curriculum of tasks with progressing difficulty for evaluating continual learning. Additional details and demonstrations for this system can be found at https://plan-with-climb.github.io/ .
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。