Optimal Coverage Path Planning on 3D Terrain

Abstract

Automated path planning is important for the optimization of field operations. Field operations should be done in a manner that minimizes time, travels over the field surface and are coordinated with specific field operations and topographic land features. Intelligent algorithms are desired for both 2D and 3D terrain field coverage path planning. The algorithm of generating an optimized full coverage pattern for a given 2D planar field by using boustrophedon paths has been investigated and reported before. However, in real world, a great proportion of farms have rolling terrains, which have a considerable influence to the design of coverage paths. Coverage path planning in 3D space has a great potential to further optimize field operations. To achieve this goal, following four research tasks are critical: terrain modeling and representation; coverage cost analysis; terrain decomposition; and the development of optimal path searching algorithm. This work addressed these four tasks and the developed algorithms and methods have been successfully implemented and tested using 3D terrain maps of farm fields with various topographic features. Each field was decomposed into sub-regions based on its terrain features. A recommended “seed curve” based on a customized cost function was searched for each sub-region and parallel coverage paths were generated by offsetting the found “seed curve” sideways until the whole region was completely covered. Compared with the 2D planning results, the experimental results of 3D coverage path planning showed its superiority in reducing both headland turning cost and soil erosion cost. On the tested fields, on average the 3D planning algorithm saved 10.3% on headland turning cost, 24.7% on soil erosion cost, 81.2% on skipped area cost, and 22.0% on the weighted sum of these costs, where their corresponding weights were 1, 1, 0.5, respectively.