Learning long‐range vision for autonomous off‐road driving

By Raia Hadsell, Pierre Sermanet, Jan Ben, Ayse Erkan, Marco Scoffier, Koray Kavukcuoglu, Urs Muller, Yann LeCun. 2009

320 citations recorded by the source corpus.

Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition, Robotics & Control

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Aerospace Engineering

In Computer Vision and Pattern Recognition

In Robotics & Control

Citations in the corpus

Self-Supervised Deep Reinforcement Learning with Generalized Computation Graphs for Robot Navigation (2017, methodology): “These supervised learning methods include learning: drivable routes and then using a planner [13], near-to-far obstacle detectors [14], reactive controllers on top of a map-based planner [15], driving affor-”
BADGR: An Autonomous Self-Supervised Learning-Based Navigation System (2020, neutral): “However,thesemethodstypicallydependonexistingSLAM approaches[5],[11]–[13]orhumans[14],[15]inorder toprovidethesemanticlabels,whichconsequentlymeans theseapproacheseitherinheritthelimitationsofgeometric approachesorarenotautonomouslyself-improving.”
Composable Action-Conditioned Predictors: Flexible Off-Policy Learning for Robot Navigation (2018, background): “Supervised learning for navigation using off-policy data has been investigated, including learning drivable routes [24] and near-to-far obstacle detectors [25].”
Deep visual foresight for planning robot motion (2016, background): “Several works have used deep neural networks to process images and represent policies for robotic control, initially for driving tasks [15], [16], later for robotic soccer [17], and most recently for robotic grasping [18], [19] and manipulation [20].”
DeepDriving: Learning Affordance for Direct Perception in Autonomous Driving (2015, background): “[9] proposes an off-road driving robot with self-supervised learning ability for long-range vision.”
Learning monocular reactive UAV control in cluttered natural environments (2012, background): “Michels et al. [22] demonstrated driving a remote-controlled toy car outdoors using monocular vision and reinforcement learning.”
Cognitive Mapping and Planning for Visual Navigation (2017, background): “Learning based approaches (Zamir et al. 2016; Hadsell et al. 2009) study the problem in isolation thus only learning generic task-independent maps.”
Large-Scale FPGA-based Convolutional Networks (2011, methodology): “In Hadsell et al. (2009), the system is pre-trained off-line using a combination of unsupervised learning (as described in section 1.3) and supervised learning.”
NeuFlow: A runtime reconfigurable dataflow processor for vision (2011, methodology): “Multiple object detection [21] or online learning for adaptive robot guidance [13] are tasks that are currently being developed around neuFlow, using and extending the convolutional network framework.”
Hardware accelerated convolutional neural networks for synthetic vision systems (2010, background): “Multiple objec t detection [5] or online learning for adaptive robot guidanc e [9] are tasks that will be largely improved by this system.”
EBLearn: Open-Source Energy-Based Learning in C++ (2009, background): “Inference consists in finding configurations of output and latent variables that minimize the energy.”
Student-teacher network learning with enhanced features (2017, methodology): “The use of a teacher model based on more accurate sensor information is also known as self-supervised learning [14, 15].”
RoboTHOR: An Open Simulation-to-Real Embodied AI Platform (2020, background): “Spatial navigation approaches [62, 17, 64, 23, 25, 56, 79, 49, 30, 14, 10] typically address navigating towards a pre-specified coordinate or a frame of a scene and they focus on understanding the geometry of the scene and learning better exploration strategies.”
Aggressive Deep Driving: Model Predictive Control with a CNN Cost Model (2017, background): “[17] learns a neural network to predict far-field traversibility from images, which is then fed into a separate planning and control framework.”
An Introduction to Deep Learning (2011, unknown): “Deep networks have been largely applied to visual classiﬁcation databases such as handwritten digits 1 , object categories 2 3 4 , pedestrian detection [33] or oﬀ-road robot navigation [34], and also on acoustic signals to perform audio clas-siﬁcation [35].”
FPGA-Based Accelerators of Deep Learning Networks for Learning and Classification: A Review (2019, background): “…or unsafe zones), medical research (e.g., in identiﬁcation of cancer cells), industrial automation (e.g., to improve worker safety by detecting when people or objects are within an unsafe distance of machines), and electronics (used in automated hearing, speech translation, etc.) [9], [43]–[46].”