Navigational Path-Planning For All-Terrain Autonomous Agricultural Robot

TL;DR

This paper compares novel path planning algorithms for autonomous agricultural robots navigating complex farmlands, emphasizing high-resolution grid maps tailored to Indian environments to improve efficiency and robustness.

Contribution

It introduces a grid map-based approach for farmland navigation and evaluates multiple algorithms' performance in realistic simulation settings.

Findings

Algorithms are effective for autonomous navigation in farmland environments.

High-resolution grid maps improve obstacle detection and path accuracy.

The methods demonstrate robustness to environmental changes.

Abstract

The shortage of workforce and increasing cost of maintenance has forced many farm industrialists to shift towards automated and mechanized approaches. The key component for autonomous systems is the path planning techniques used. Coverage path planning (CPP) algorithm is used for navigating over farmlands to perform various agricultural operations such as seeding, ploughing, or spraying pesticides and fertilizers. This report paper compares novel algorithms for autonomous navigation of farmlands. For reduction of navigational constraints, a high-resolution grid map representation is taken into consideration specific to Indian environments. The free space is covered by distinguishing the grid cells as covered, unexplored, partially explored and presence of an obstacle. The performance of the compared algorithms is evaluated with metrics such as time efficiency, space efficiency, accuracy, and robustness to changes in the environment. Robotic Operating System (ROS), Dassault Systemes Experience Platform (3DS Experience), MATLAB along Python were used for the simulation of the compared algorithms. The results proved the applicability of the algorithms for autonomous field navigation and feasibility with robotic path planning.