GPU-Assisted Learning on an Autonomous Marine Robot for Vision-Based Navigation and Image Understanding

We present a GPU-based integrated robotic platform that enables collision avoidance, navigation, and image understanding on a single underwater vehicle. The platform enables observational tasks such as coral reef health assessment by enabling simultaneous operation of multiple image analysis taskswhile navigating in close proximity to obstacles. The integration of a GPU allows us to leverage deep neural networks for collision avoidance and automated object detection and classification while a general purpose CPU processes images to perform visual Simultaneous Localization and Mapping (SLAM). In this paper, we describe the system architecture and summarize experimental results for coral detection and collision-free navigation.

[1]  Nikolai Smolyanskiy,et al.  Toward low-flying autonomous MAV trail navigation using deep neural networks for environmental awareness , 2017, 2017 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

[2]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[3]  Gregory Dudek,et al.  Towards Autonomous Robotic Coral Reef Health Assessment , 2015, FSR.

[4]  Gregory Dudek,et al.  Vision-Based Autonomous Underwater Swimming in Dense Coral for Combined Collision Avoidance and Target Selection , 2018, 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

[5]  Gregory Dudek,et al.  Wide-Speed Autopilot System for a Swimming Hexapod Robot , 2013, 2013 International Conference on Computer and Robot Vision.

[6]  Lino Marques,et al.  Robots for Environmental Monitoring: Significant Advancements and Applications , 2012, IEEE Robotics & Automation Magazine.

[7]  Dean Pomerleau,et al.  ALVINN, an autonomous land vehicle in a neural network , 2015 .

[8]  Peter I. Corke,et al.  Experiments with Underwater Robot Localization and Tracking , 2007, Proceedings 2007 IEEE International Conference on Robotics and Automation.

[9]  Andrew Hogue,et al.  A visually guided swimming robot , 2005, 2005 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[10]  Gregory Dudek,et al.  Synthetically Trained 3D Visual Tracker of Underwater Vehicles , 2018, OCEANS 2018 MTS/IEEE Charleston.

[11]  R. Davis,et al.  The autonomous underwater glider "Spray" , 2001 .

[12]  Yuan Yu,et al.  TensorFlow: A system for large-scale machine learning , 2016, OSDI.

[13]  Bo He,et al.  Stereo visual SLAM system in underwater environment , 2014, OCEANS 2014 - TAIPEI.

[14]  Daniel Toal,et al.  Computer Vision Applications in the Navigation of Unmanned Underwater Vehicles , 2009 .

[15]  Carlos R. del-Blanco,et al.  DroNet: Learning to Fly by Driving , 2018, IEEE Robotics and Automation Letters.

[16]  Dana R. Yoerger,et al.  Navigation and control of the Nereus hybrid underwater vehicle for global ocean science to 10,903 m depth: Preliminary results , 2010, 2010 IEEE International Conference on Robotics and Automation.

[17]  Xin Zhang,et al.  End to End Learning for Self-Driving Cars , 2016, ArXiv.

[18]  Ryan M. Eustice,et al.  Perception-driven navigation: Active visual SLAM for robotic area coverage , 2013, 2013 IEEE International Conference on Robotics and Automation.

[19]  C. C. Eriksen,et al.  Seaglider: a long-range autonomous underwater vehicle for oceanographic research , 2001 .

[20]  Gregory Dudek,et al.  3D trajectory synthesis and control for a legged swimming robot , 2014, 2014 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[21]  J. M. M. Montiel,et al.  ORB-SLAM: A Versatile and Accurate Monocular SLAM System , 2015, IEEE Transactions on Robotics.

[22]  Daniel Cremers,et al.  Direct Sparse Odometry , 2016, IEEE Transactions on Pattern Analysis and Machine Intelligence.