CPP-DIP: Multi-objective Coverage Path Planning for MAVs in Dispersed and Irregular Plantations (2505.04989v1)

Abstract: Coverage Path Planning (CPP) is vital in precision agriculture to improve efficiency and resource utilization. In irregular and dispersed plantations, traditional grid-based CPP often causes redundant coverage over non-vegetated areas, leading to waste and pollution. To overcome these limitations, we propose CPP-DIP, a multi-objective CPP framework designed for Micro Air Vehicles (MAVs). The framework transforms the CPP task into a Traveling Salesman Problem (TSP) and optimizes flight paths by minimizing travel distance, turning angles, and intersection counts. Unlike conventional approaches, our method does not rely on GPS-based environmental modeling. Instead, it uses aerial imagery and a Histogram of Oriented Gradients (HOG)-based approach to detect trees and extract image coordinates. A density-aware waypoint strategy is applied: Kernel Density Estimation (KDE) is used to reduce redundant waypoints in dense regions, while a greedy algorithm ensures complete coverage in sparse areas. To verify the generality of the framework, we solve the resulting TSP using three different methods: Greedy Heuristic Insertion (GHI), Ant Colony Optimization (ACO), and Monte Carlo Reinforcement Learning (MCRL). Then an object-based optimization is applied to further refine the resulting path. Additionally, CPP-DIP integrates ForaNav, our insect-inspired navigation method, for accurate tree localization and tracking. The experimental results show that MCRL offers a balanced solution, reducing the travel distance by 16.9 % compared to ACO while maintaining a similar performance to GHI. It also improves path smoothness by reducing turning angles by 28.3 % and 59.9 % relative to ACO and GHI, respectively, and effectively eliminates intersections. These results confirm the robustness and effectiveness of CPP-DIP in different TSP solvers.

Multi-objective Coverage Path Planning for MAVs in Dispersed and Irregular Plantations

Precision agriculture is undergoing a transformation with the advent of micro air vehicles (MAVs), which offer unparalleled efficiency and adaptability in diverse environments. The research paper titled "CPP-DIP: Multi-objective Coverage Path Planning for MAVs in Dispersed and Irregular Plantations" introduces a novel framework for addressing the complexities inherent in coverage path planning (CPP) within dispersed and irregular plantations, a context not typically considered in traditional grid-based approaches.

Objectives and Methodology

The CPP-DIP framework aims to optimize multiple objectives simultaneously: minimizing travel distance, reducing turning angles, and eliminating path intersections. This multi-objective approach ensures efficient coverage while conserving energy and reducing environmental impact. This paper implements a task transformation akin to the Traveling Salesman Problem (TSP), optimizing flight routes without reliance on GPS-based modeling, which is often costly and time-consuming in large plantation contexts.

Detection and Path Planning

Central to the framework is an image-based system that employs aerial imagery coupled with Histogram of Oriented Gradients (HOG)-based object detection to identify tree positions. This approach eschews traditional GPS methods, offering a streamlined and cost-effective alternative. The detected tree coordinates serve as foundational data points for the CPP task, where Kernel Density Estimation (KDE) aids in streamlining waypoints in densely packed regions. A greedy algorithm ensures sparse areas receive adequate coverage.

Robustness and Comparisons

The researchers tested the CPP-DIP framework using three TSP solvers: Greedy Heuristic Insertion (GHI), Ant Colony Optimization (ACO), and Monte Carlo Reinforcement Learning (MCRL). Notably, MCRL emerged as a balanced solution across various metrics. According to experimental results, MCRL reduced travel distance by 16.9% compared to ACO, while significantly enhancing path smoothness, demonstrated by a reduction in turning angles by 28.3% compared to ACO and 59.9% compared to GHI. Moreover, MCRL was adept at eliminating intersections, a notable advantage over other methods.

Practical Implications

The implications of this research are profound, offering a framework that can be applied to optimize MAV usage in agricultural environments where plantation density and arrangement are non-uniform. The decline in redundant coverage represents a direct enhancement in resource utilization—less pesticide overspraying and reduced chemical waste, mitigating pollution and economic costs.

Theoretical Impact and Future Directions

Theoretically, this research sets a precedent in transforming CPP tasks using non-GPS technologies, highlighting the potential for image-based navigation systems in robotics and automation sectors. Future advancements may explore real-time integration of such systems, enabling MAVs to adapt dynamically to environmental changes during flight. The inclusion of multi-agent systems could lead to even more efficient coverage, expanding the scale and applicability of such frameworks in broader agricultural settings.

Conclusion

The CPP-DIP research significantly advances coverage path planning by providing a robust framework for MAVs in complex plantation environments. By balancing multiple objectives effectively, it addresses critical concerns of efficiency, precision, and environmental stewardship. This innovative approach stands to reshape methodologies in precision agriculture, offering a sustainable model adaptable to various real-world scenarios.