Evaluating Path Planning Strategies for Efficient Nitrate Sampling in Crop Rows

2025-03-10

AI Summary

This work addresses the path-planning challenge for autonomous ground robots navigating between crop rows during precision nitrate sampling in agricultural fields. We propose a navigation framework integrating UAV-derived orthomosaic maps with high-precision GPS. For the first time, we systematically benchmark three planning paradigms—A*, Deep Q-Network (DQN), and heuristic search—under strong geometric constraints imposed by narrow, structured crop-row environments. Results show that the heuristic approach achieves optimal performance: 0.28 ms average planning latency and 100% task success rate; A* performs nearly as well; DQN suffers from inference latency and local suboptimality, limiting its practicality. The study demonstrates the clear advantages of deterministic, rule-based methods in real-time responsiveness, robustness, and deployment feasibility. It establishes a reproducible benchmarking methodology and provides empirically grounded guidance for algorithm selection in narrow-field agricultural robot navigation.

Technology Category

• Intelligent Robots: Motion and Path Planning
• Search and Optimization: Heuristic Search
• Planning, Routing, and Scheduling: Deterministic Planning

Application Category

• Search and Retrieval-Augmented AI: Agentic search
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web
• Systems and Infrastructure for Web, Mobile and WoT: Experiences and lessons learnt from Web-based algorithms and system deployments

Abstract

This paper presents a pipeline that combines high-resolution orthomosaic maps generated from UAS imagery with GPS-based global navigation to guide a skid-steered ground robot. We evaluated three path planning strategies: A* Graph search, Deep Q-learning (DQN) model, and Heuristic search, benchmarking them on planning time and success rate in realistic simulation environments. Experimental results reveal that the Heuristic search achieves the fastest planning times (0.28 ms) and a 100% success rate, while the A* approach delivers near-optimal performance, and the DQN model, despite its adaptability, incurs longer planning delays and occasional suboptimal routing. These results highlight the advantages of deterministic rule-based methods in geometrically constrained crop-row environments and lay the groundwork for future hybrid strategies in precision agriculture.

Problem

• Evaluates path planning strategies for nitrate sampling in crop rows.
• Compares A*, DQN, and Heuristic search for efficiency and success rate.
• Demonstrates Heuristic search's superiority in speed and success rate.

Innovation

• Combines UAS imagery with GPS navigation
• Evaluates A*, DQN, and Heuristic path planning
• Heuristic search achieves fastest planning times