ATLASv2: LLM-Guided Adaptive Landmark Acquisition and Navigation on the Edge
This addresses the challenge of deploying autonomous systems on edge devices with real-time processing demands, though it appears incremental as an improved version of a prior system.
The paper tackles the problem of autonomous navigation and manipulation on resource-constrained edge devices by introducing ATLASv2, which integrates a fine-tuned TinyLLM, real-time object detection, and path planning to achieve high success rates in real-world home and office environments with minimal latency and power consumption.
Autonomous systems deployed on edge devices face significant challenges, including resource constraints, real-time processing demands, and adapting to dynamic environments. This work introduces ATLASv2, a novel system that integrates a fine-tuned TinyLLM, real-time object detection, and efficient path planning to enable hierarchical, multi-task navigation and manipulation all on the edge device, Jetson Nano. ATLASv2 dynamically expands its navigable landmarks by detecting and localizing objects in the environment which are saved to its internal knowledge base to be used for future task execution. We evaluate ATLASv2 in real-world environments, including a handcrafted home and office setting constructed with diverse objects and landmarks. Results show that ATLASv2 effectively interprets natural language instructions, decomposes them into low-level actions, and executes tasks with high success rates. By leveraging generative AI in a fully on-board framework, ATLASv2 achieves optimized resource utilization with minimal prompting latency and power consumption, bridging the gap between simulated environments and real-world applications.