CNN-based single image obstacle avoidance on a quadrotor

This paper demonstrates the use of a single forward facing camera for obstacle avoidance on a quadrotor. We train a CNN for estimating depth from a single image. The depth map is then fed to a behaviour arbitration based control algorithm that steers the quadrotor away from obstacles. We conduct experiments with simulated and real drones in a variety of environments.

[1]  Ashutosh Saxena,et al.  High speed obstacle avoidance using monocular vision and reinforcement learning , 2005, ICML.

[2]  Ian D. Reid,et al.  Learning Depth from Single Monocular Images Using Deep Convolutional Neural Fields , 2015, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[3]  Ashutosh Saxena,et al.  Learning Depth from Single Monocular Images , 2005, NIPS.

[4]  Sergio Montenegro,et al.  Obstacle Detection and Collision Avoidance for a UAV With Complementary Low-Cost Sensors , 2015, IEEE Access.

[5]  Martial Hebert,et al.  Learning Transferable Policies for Monocular Reactive MAV Control , 2016, ISER.

[6]  Ashutosh Saxena,et al.  Low-power parallel algorithms for single image based obstacle avoidance in aerial robots , 2012, 2012 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[7]  Ashutosh Saxena,et al.  Autonomous MAV flight in indoor environments using single image perspective cues , 2011, 2011 IEEE International Conference on Robotics and Automation.

[8]  Martial Hebert,et al.  Robust Monocular Flight in Cluttered Outdoor Environments , 2016, ArXiv.

[9]  Daniel Cremers,et al.  Collision Avoidance for Quadrotors with a Monocular Camera , 2014, ISER.

[10]  Andrea Vedaldi,et al.  MatConvNet: Convolutional Neural Networks for MATLAB , 2014, ACM Multimedia.

[11]  Paolo Valigi,et al.  Fast robust monocular depth estimation for Obstacle Detection with fully convolutional networks , 2016, 2016 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

[12]  Jürgen Schmidhuber,et al.  A Machine Learning Approach to Visual Perception of Forest Trails for Mobile Robots , 2016, IEEE Robotics and Automation Letters.

[13]  Henrik I. Christensen,et al.  Behaviour coordination for navigation in office environments , 2002, IEEE/RSJ International Conference on Intelligent Robots and Systems.

[14]  Marc Pollefeys,et al.  Reactive avoidance using embedded stereo vision for MAV flight , 2015, 2015 IEEE International Conference on Robotics and Automation (ICRA).

[15]  Rob Fergus,et al.  Depth Map Prediction from a Single Image using a Multi-Scale Deep Network , 2014, NIPS.

[16]  Russ Tedrake,et al.  Pushbroom stereo for high-speed navigation in cluttered environments , 2014, 2015 IEEE International Conference on Robotics and Automation (ICRA).