Perception-Based UAV Fruit Grasping Using Sub-Task Imitation Learning
Perception-Based UAV Fruit Grasping Using Sub-Task Imitation Learning
复制标题
DOI：
10.1109/airpharo52252.2021.9571066
复制
发表时间：
2021-10
期刊：
影响因子：
--
通讯作者：
Gabriel Baraban;Siddharth Kothiyal;Marin Kobilarov
中科院分区：
文献类型：
--
作者：
Gabriel Baraban;Siddharth Kothiyal;Marin Kobilarov
这项工作考虑了通过将基于视觉的感知和控制紧密地集成在学习框架内，使用空中抓取机器人来自主采摘水果。该体系结构采用卷积神经网络(CNN)对图像和车辆状态信息进行编码。该编码被传递给子任务分类器和相关联的参考路点生成器。训练分类器以预测正在执行的任务的当前阶段：暂存、挑选或重置。根据预测的相位，航点生成器预测一组无障碍的六自由度航点，作为模型预测控制(MPC)的参考轨迹。通过迭代地生成并跟踪这些轨迹，空中操纵器安全地接近模拟目标水果并将其从树上移除。通过与常规基线方法的比较，以及对其关键特性的烧蚀研究，该方法在29次飞行试验中得到了验证。总体而言，该方法取得了与传统方法相当的成功率，同时更快地达到目标。
This work considers autonomous fruit picking using an aerial grasping robot by tightly integrating vision-based perception and control within a learning framework. The architecture employs a convolutional neural network (CNN) to encode images and vehicle state information. This encoding is passed into a sub-task classifier and associated reference waypoint generator. The classifier is trained to predict the current phase of the task being executed: Staging, Picking, or Reset. Based on the predicted phase, the waypoint generator predicts a set of obstacle-free 6-DOF waypoints, which serve as a reference trajectory for model-predictive control (MPC). By iteratively generating and following these trajectories, the aerial manipulator safely approaches a mock-up goal fruit and removes it from the tree. The proposed approach is validated in 29 flight tests, through a comparison to a conventional baseline approach, and an ablation study on its key features. Overall, the approach achieved comparable success rates to the conventional approach, while reaching the goal faster.