Position-Agnostic Autonomous Navigation in Vineyards with Deep Reinforcement Learning

被引:20
作者
Martini, Mauro [1 ,2 ]
Cerrato, Simone [1 ,2 ]
Salvetti, Francesco [1 ,2 ,3 ]
Angarano, Simone [1 ,2 ]
Chiaberge, Marcello [1 ,2 ]
机构
[1] Politecn Torino, Dept Elect & Telecommun, Turin, Italy
[2] Politecn Torino, PIC4SeR Interdept Ctr Serv Robot, Turin, Italy
[3] Politecn Torino, SmartData Interdept Ctr Big Data & Data Sci, Turin, Italy
来源
2022 IEEE 18TH INTERNATIONAL CONFERENCE ON AUTOMATION SCIENCE AND ENGINEERING (CASE) | 2022年
关键词
AGRICULTURE;
D O I
10.1109/CASE49997.2022.9926582
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Precision agriculture is rapidly attracting research to efficiently introduce automation and robotics solutions to support agricultural activities. Robotic navigation in vineyards and orchards offers competitive advantages in autonomously monitoring and easily accessing crops for harvesting, spraying and performing time-consuming necessary tasks. Nowadays, autonomous navigation algorithms exploit expensive sensors which also require heavy computational cost for data processing. Nonetheless, vineyard rows represent a challenging outdoor scenario where GPS and Visual Odometry techniques often struggle to provide reliable positioning information. In this work, we combine Edge AI with Deep Reinforcement Learning to propose a cutting-edge lightweight solution to tackle the problem of autonomous vineyard navigation without exploiting precise localization data and overcoming task-tailored algorithms with a flexible learning-based approach. We train an end-to-end sensorimotor agent which directly maps noisy depth images and position-agnostic robot state information to velocity commands and guides the robot to the end of a row, continuously adjusting its heading for a collision-free central trajectory. Our extensive experimentation in realistic simulated vineyards demonstrates the effectiveness of our solution and the generalization capabilities of our agent.
引用
收藏
页码:477 / 484
页数:8
相关论文
共 40 条
  • [1] Deep Semantic Segmentation at the Edge for Autonomous Navigation in Vineyard Rows
    Aghi, Diego
    Cerrato, Simone
    Mazzia, Vittorio
    Chiaberge, Marcello
    [J]. 2021 IEEE/RSJ INTERNATIONAL CONFERENCE ON INTELLIGENT ROBOTS AND SYSTEMS (IROS), 2021, : 3421 - 3428
  • [2] Local Motion Planner for Autonomous Navigation in Vineyards with a RGB-D Camera-Based Algorithm and Deep Learning Synergy
    Aghi, Diego
    Mazzia, Vittorio
    Chiaberge, Marcello
    [J]. MACHINES, 2020, 8 (02)
  • [3] Monocular Visual Odometry Using Fisheye Lens Cameras
    Aguiar, Andre
    Santos, Filipe
    Santos, Luis
    Sousa, Armando
    [J]. PROGRESS IN ARTIFICIAL INTELLIGENCE, PT II, 2019, 11805 : 319 - 330
  • [4] Survey of Deep Reinforcement Learning for Motion Planning of Autonomous Vehicles
    Aradi, Szilard
    [J]. IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS, 2022, 23 (02) : 740 - 759
  • [5] Harvesting Robots for High-value Crops: State-of-the-art Review and Challenges Ahead
    Bac, C. Wouter
    van Henten, Eldert J.
    Hemming, Jochen
    Edan, Yael
    [J]. JOURNAL OF FIELD ROBOTICS, 2014, 31 (06) : 888 - 911
  • [6] Detection, classification, and mapping of coffee fruits during harvest with computer vision
    Bazame, Helizani Couto
    Molin, Jose Paulo
    Althoff, Daniel
    Martello, Mauricio
    [J]. COMPUTERS AND ELECTRONICS IN AGRICULTURE, 2021, 183
  • [7] Grape clusters and foliage detection algorithms for autonomous selective vineyard sprayer
    Berenstein, Ron
    Ben Shahar, Ohad
    Shapiro, Amir
    Edan, Yael
    [J]. INTELLIGENT SERVICE ROBOTICS, 2010, 3 (04) : 233 - 243
  • [8] Callegati F., 2018, WORKSHOP SMALL UAVS
  • [9] Cerrato S, 2023, Arxiv, DOI [arXiv:2112.03816, DOI 10.48550/ARXIV.2112.03816]
  • [10] Chen YF, 2017, IEEE INT C INT ROBOT, P1343, DOI 10.1109/IROS.2017.8202312