In-field path planning method for soil compaction reduction by returning soil to the field
By dividing the fields using multispectral remote sensing and deep learning models, and combining particle swarm optimization algorithms to plan the path of the fertilization machine, the problem of neglecting soil compaction in traditional fertilization machine path planning is solved. This reduces the area of furrow compaction and the number of repeated compaction cycles, thereby improving the soil improvement effect.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- NORTHEAST AGRICULTURAL UNIVERSITY
- Filing Date
- 2025-09-26
- Publication Date
- 2026-05-26
AI Technical Summary
Traditional straw return machines neglect the cumulative effect of soil compaction in their path planning, which causes the soil improvement effect of straw return to the field to be partially offset or even exacerbate the risk of soil degradation.
Multispectral remote sensing technology and deep learning models were used to divide the fields into plots. Particle swarm optimization algorithm was combined to plan the path of the fertilization machine to reduce the area of furrow compaction and the number of repeated compaction times. The path was optimized by objective function to ensure the shortest path while reducing soil structure damage.
While ensuring the efficiency of straw return to the field, we aim to minimize the damage to soil structure caused by mechanical operations, achieve the dual goals of "returning straw to the field" and "preserving soil," and provide technical support for smart agriculture.
Smart Images

Figure CN121140829B_ABST