# Agentic Self-Evolutionary Replanning for Embodied Navigation

arXiv: https://arxiv.org/abs/2603.02772 · HTML: https://mlsanity.com/paper/2603.02772/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Placed in Multimodal and embodied agents, one of 16 directions in LLM agents and tool use.
- Builds on 21 of its 29 references that are in the map.
- First posted on arXiv 3 Mar 2026.

## Placed in

- LLM agents and tool use / [Multimodal and embodied agents](https://mlsanity.com/agents/multimodal-and-embodied-agents/)

## Builds on (most cited first, 21 of 21)

- [Reflexion: Language Agents with Verbal Reinforcement Learning](https://mlsanity.com/paper/2303.11366/) (2023)
- [Voyager: An Open-Ended Embodied Agent with Large Language Models](https://mlsanity.com/paper/2305.16291/) (2023)
- [ExpeL: LLM Agents Are Experiential Learners](https://mlsanity.com/paper/2308.10144/) (2023)
- [Habitat-Matterport 3D Dataset (HM3D): 1000 Large-scale 3D Environments for Embodied AI](https://mlsanity.com/paper/2109.08238/) (2021)
- [Eureka: Human-Level Reward Design via Coding Large Language Models](https://mlsanity.com/paper/2310.12931/) (2023)
- [Differentiable Convex Optimization Layers](https://mlsanity.com/paper/1910.12430/) (2019)
- [JARVIS-1: Open-World Multi-task Agents with Memory-Augmented Multimodal Language Models](https://mlsanity.com/paper/2311.05997/) (2023)
- [Text2Reward: Reward Shaping with Language Models for Reinforcement Learning](https://mlsanity.com/paper/2309.11489/) (2023)
- [DrEureka: Language Model Guided Sim-To-Real Transfer](https://mlsanity.com/paper/2406.01967/) (2024)
- [ReMEmbR: Building and Reasoning Over Long-Horizon Spatio-Temporal Memory for Robot Navigation](https://mlsanity.com/paper/2409.13682/) (2024)
- [Motion Planning Networks: Bridging the Gap Between Learning-based and Classical Motion Planners](https://mlsanity.com/paper/1907.06013/) (2019)
- [Open-Ended Instructable Embodied Agents with Memory-Augmented Large Language Models](https://mlsanity.com/paper/2310.15127/) (2023)
- [Optimization-Based Collision Avoidance](https://mlsanity.com/paper/1711.03449/) (2017)
- [RAPTOR: Robust and Perception-aware Trajectory Replanning for Quadrotor Fast Flight](https://mlsanity.com/paper/2007.03465/) (2020)
- [RDA: An Accelerated Collision Free Motion Planner for Autonomous Navigation in Cluttered Environments](https://mlsanity.com/paper/2210.00192/) (2022)
- [NeuPAN: Direct Point Robot Navigation with End-to-End Model-based Learning](https://mlsanity.com/paper/2403.06828/) (2024)
- [Teach-Repeat-Replan: A Complete and Robust System for Aggressive Flight in Complex Environments](https://mlsanity.com/paper/1907.00520/) (2019)
- [MSI-Agent: Incorporating Multi-Scale Insight into Embodied Agents for Superior Planning and Decision-Making](https://mlsanity.com/paper/2409.16686/) (2024)
- [LERa: Replanning with Visual Feedback in Instruction Following](https://mlsanity.com/paper/2507.05135/) (2025)
- [Towards real-world navigation with deep differentiable planners](https://mlsanity.com/paper/2108.05713/) (2021)
- [DiffTune-MPC: Closed-Loop Learning for Model Predictive Control](https://mlsanity.com/paper/2312.11384/) (2023)

## Built on by (most cited first, 2 of 2)

- [The Last AI Built by Humans: Toward Genuine Recursive Self-Improvement](https://mlsanity.com/paper/2609.11873/) (2026)
- [ROSClaw: A Hierarchical Semantic-Physical Framework for Heterogeneous Multi-Agent Collaboration](https://mlsanity.com/paper/2604.04664/) (2026)

## Abstract (arXiv)

Motion planners tuned under nominal conditions can fail when execution conditions change. Existing differentiable adaptation methods provide efficient online updates but remain constrained by static loss functions and gradient-only parameter updates. We propose Self-Evolutionary RePlanning (SERP), a failure-aware framework that jointly adapts motion-planner parameters and loss function weights from closed-loop execution feedback. SERP provides two adaptation mechanisms. First, in-context learning with auto-differentiation (ILAD) enables rapid online adaptation by coupling local numerical refinement with history-conditioned semantic intervention. Auto-differentiation performs numerical updates, while a vision-language model (VLM) interprets recent adaptation history to jointly revise planner parameters and loss weights when local adaptation becomes ineffective. Second, the memory-augmented rollout system (MARS) extends ILAD into a deliberative and evolutionary recovery process with memory-guided candidate generation and parallel low-fidelity rollout evaluation. The resulting recovery trace is then consolidated into recovery memory to guide future adaptation. Extensive simulation and real-world experiments demonstrate improved failure recovery performance over various baselines across diverse environments.