Authors

Abstract

In recent years, the development of vineyard robots has emerged as a significant development in agricultural equipment, playing an increasingly vital role in precision agriculture and intelligent operations. These robots are capable of precise navigation, obstacle avoidance, and real-time path planning within agricultural settings. The paper employs laser Simultaneous Localization and Mapping (SLAM) technology as the primary method for achieving real-time, accurate positioning of the robot, thereby providing reliable environmental perception capabilities and prior map information for the vineyard robot. The Robust-Time Elastic Band (R-TEB) local planning algorithm developed in this study automatically generates a smooth, continuous inspection path within the operational area. This objective is pursued by the consideration of parameters such as the robot's working width, minimum turning radius, and operational strip width, with the aim of achieving a minimization of energy consumption. Utilizing the Root Mean Square Error (RMSE) metric to gauge prediction accuracy, the R-TEB algorithm yielded values ranging from 0.016 to 0.022 meters, while the TEB algorithm produced values between 0.012 and 0.025 meters. The findings indicate that the R-TEB algorithm optimizes trajectory quality in vineyard environments, thereby enhancing the robot's autonomous navigation capabilities and obstacle avoidance efficiency.

Abstract in Chinese

近年来，葡萄园机器人的研发已成为农业设备领域的重要突破，在精准农业与智能化作业中发挥着日益关键的作用。这类机器人能够在农业场景中实现精准导航、障碍物规避及实时路径规划。本文采用激光同步定位与建图（SLAM）技术作为核心方法，实现机器人实时精准定位，从而为葡萄园机器人提供可靠的环境感知能力和预先地图信息。研究开发的鲁棒时间弹性带（R-TEB）局部规划算法，能在作业区域内自动生成平滑连续的巡检路径。该算法通过综合考虑机器人作业宽度、最小转弯半径及作业带宽度等参数，实现能耗最小化目标。采用均方根误差（RMSE）指标评估预测精度时，R-TEB算法得出0.016至0.022米的误差值，而TEB算法误差值介于0.012至0.025米之间。研究表明，R-TEB算法能优化葡萄园环境中的轨迹质量，从而提升机器人的自主导航能力和避障效率。