Large Language Models for Task Planning in Embodied AI: A Survey

Large Language Model-Driven Symbolic Planning for Long-Horizon Robotic Manipulation Tasks

Cross-modal foundation models are increasingly used for robotic task understanding and planning. However, connecting multimodal observations and natural language instructions to symbolic planners and executable robot actions remains a key challenge in embodied AI. In addressing this challenge, existing methods often rely on structured inputs and lack robustness to language ambiguity, with limited real-world validation. We propose VLA-SP (Vision-Language-Action via Symbolic Planning), a two-stage Embodied Vision-Language-Action framework. VLA-SP consists of a Multi-modal Information Processing Module and a PDDL-based Vision-Language-Action Module, enabling fully automated robotic execution from speech and vision inputs. First, our framework detects task-relevant objects using YOLO and transcribes voice instructions with Faster-Whisper. After obtaining the visual observations and language instructions, the LLM fuses these multimodal inputs and converts them into a planner-verifiable problem.pddl instance under a reusable domain.pddl. Then, Fast Downward validates the generated instance and, upon successful verification, produces a multi-step symbolic plan. If any issues are detected during verification, planner feedback is used for traceable failure diagnosis and iterative self-correction. Once a valid symbolic plan is obtained, the symbolic actions are further grounded into 6-DoF poses and executed through predefined robot skills. Finally, we conduct extensive experiments across seven planning domains in simulation and six real-world manipulation tasks on two robotic platforms, X-Arm and Franka. The experimental results demonstrate that VLA-SP enables reliable execution with high success rates across diverse manipulation tasks. In particular, the complete framework achieves an average success rate of 82% across the six real-world tasks. Results from both simulated and real-world experiments fur-ther demonstrate the strong interpretability, executability, and cross-platform applicability of our framework. Further details are available here.

Plan Right, Then Plan Tight: Symbolic RL for Efficient Embodied Reasoning

Embodied task planning asks an agent to turn a natural-language instruction into an executable sequence of actions in a physical scene, and is a building block for household, assistive, and service robots. Recent prompting-based and reinforcement-learning planners generate fluent action text but lack a cheap deterministic check that the produced plan is valid in the target world, while high-fidelity simulation is too slow to serve as an inner-loop training signal. The general problem is therefore how to obtain verifiable supervision and rewards for embodied planners without relying on string-level matching or full simulation. Here we show that a single BDDL specification, automatically constructed from open-world video evidence or curated tasks, can serve as a shared interface for data construction, plan verification, and reward design. A video-to-BDDL parser, an LLM verifier, and a lightweight symbolic engine together supply dense feedback at millisecond latency. We further introduce GroupAdapt, a difficulty-aware length schedule that uses the in-batch group pass rate as a zero-cost signal so that hard prompts get wider length tolerance and automatically tighten as their pass rate improves. Under the guidance of the proposed verifier and GroupAdapt schedule, the 8B planner attains a Strict-Pass score of 97.3 on BEHAVIOR-1000, yielding a 25.9 percent relative improvement over the Qwen3-8B baseline. This result exceeds the strongest large-model baseline by 3.5 percent, while simultaneously compressing the response length by 79 percent to 207 tokens, demonstrating both effectiveness and efficiency.

GroundedPlanBench: Spatially Grounded Long-Horizon Task Planning

PFEA: a VLM-based high-level natural language planning and feedback embodied agent for human-centered AI

The rapid advancement of Large Language Models (LLMs) has led to significant progress in Artificial Intelligence (AI), ushering in a new era of human-centered AI (HAI). Intelligent agents powered by LLMs provide new opportunities for realizing HAI. However, existing LLM-based embodied agents often lack online planning capabilities and may generate actions involving objects that are not present in the current environment. In this paper, we propose a closed-loop framework for planning and evaluation of a vision-language model-based robotic manipulation agent operating in tabletop object rearrangement and manipulation tasks. These tasks require a robot to interpret high-level natural language commands and perform multi-step actions such as stacking objects, sorting objects by category or attribute, placing objects into target containers, and organizing mixed desktop scenes. The proposed architecture consists of a human–robot speech interaction module, a vision–language agent module (including a planner, translator, and evaluator), and an action execution module. Within this framework, the vision–language planner decomposes high-level instructions into executable task steps via Chain-of-Objects Reasoning, enabling a deeper understanding of the visual environment, including spatial relationships and object attributes. Meanwhile, the task evaluator monitors execution outcomes and provides result-oriented feedback, allowing the system to replan when errors occur. Experimental results show that, compared with baseline methods, the proposed agent improves the average task success rate by approximately 20% in both simulation and real-world environments, significantly enhancing its ability to execute complex natural language instructions. This work demonstrates the potential of closed-loop vision-language planning for human-centered robotic manipulation. Code: https://github.com/subforpaper/PFEA.

TypeGo: An OS Runtime for Embodied Agents

Large language models (LLMs) can plan behavior for embodied agents from natural language, but treating the LLM as a request/response oracle on the critical path is fundamentally at odds with real-time control and concurrent goals. We argue for an operating-system-style runtime for embodied agents, and instantiate this idea in an early prototype, TypeGo. TypeGo structures LLM-based planning as asynchronous loops at multiple timescales that overlap with execution, and manages the agent's physical body like an OS manages hardware: the Skill Kernel arbitrates typed physical subsystems among concurrent per-task processes, a scheduler preempts them and resumes or replaces each by source, and speculative skill streaming hides LLM latency behind ongoing motion, while a fast first-action path yields visible feedback within a second. Users program behavior through natural language prescriptions that TypeGo dispatches to the LLM-based planners or compiles into low-latency interrupt handlers. Our prototype of Kalos, a Unitree Go2 quadruped, provides preliminary evidence for the design: in our current task suite, it cuts per-step delay by 50% over step-by-step planning and time-to-first-action by 73% over monolithic planning, while admitting concurrent tasks at low scheduling overhead.

Evidence-Gated Task and Motion Planning with Vision-Language Models

Robots executing long-horizon manipulation tasks from natural-language instructions must reason about both semantic task structure and geometric feasibility. However, under partial observability, the availability of goal-relevant objects may be uncertain. In such cases, approaches that combine Vision-Language Models (VLMs) with Task and Motion Planning (TAMP) may generate subgoals that rely on the VLM's prior knowledge without observational support, leading to execution failures or unintended outcomes. We propose Evidence Acquisition and Feasibility Gating (EAFG), a framework that acquires visual evidence through VLM-generated exploratory subgoals and TAMP-based execution. EAFG then applies a feasibility gate to decide whether to proceed with task planning, acquire further evidence, or halt. Our experiments show that, in cooking tasks with ambiguous object use, EAFG improves recipe completion by discovering task-relevant objects before planning. For instructions requiring an absent object, EAFG promotes appropriate halt decisions and reduces repeated attempts to manipulate that object.