Autonomous driving in urban environments: Boss and the Urban Challenge

By C. Urmson, Joshua Anhalt, J. Bagnell, Christopher R. Baker, R. Bittner, M. N. Clark, John M. Dolan, D. Duggins, Tugrul Galatali, Christopher Geyer and 32 more. 2008

1,944 citations recorded by the source corpus.

Subfields: Aerospace Engineering, Automotive 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 Automotive 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): “Although methods based on geometric reconstruction and mapping have proven effective in a range of navigation and collision avoidance domains [1], [2], [3], they impose considerable computational overhead [4] and”
Uncertainty-Aware Reinforcement Learning for Collision Avoidance (2017, background): “There is much prior work on safe robot control for safety-critical systems such as autonomous cars [29], legged robots [31], and quadrotors [20, 30, 9].”
A Survey of Autonomous Driving: Common Practices and Emerging Technologies (2019, methodology): “To better cope with the limitations and uncertainties of different sensing modalities, a sensor fusion strategy is often use for tracking [43].”
Composable Action-Conditioned Predictors: Flexible Off-Policy Learning for Robot Navigation (2018, background): “There has been extensive work on autonomous robot navigation [16, 17, 18].”
Textual Explanations for Self-Driving Vehicles (2018, neutral): “Notable examples include [23], [4], and [16].”
A Survey of Motion Planning and Control Techniques for Self-Driving Urban Vehicles (2016, methodology): “The winner of the challenge, CMU’s Boss vehicle used variational techniques for local trajectory generation in structured environments and a lattice graph in 4-dimensional configuration space (consisting of position, orientation, and velocity) together with Anytime D* to find a collision-free paths ”
Fast segmentation of 3D point clouds for ground vehicles (2010, background): “This approach was taken by many teams at the 2007 DARPA Urban Challenge robot competition in order to segment point clouds and to detect the other vehicles on the track, with great success ([3], [4], [5]).”
Toward automated driving in cities using close-to-market sensors: An overview of the V-Charge Project (2013, background): “All of the successful teams in these competitions utilized a highly sophisticated suite of expensive sensors, such as sweeping laser range ﬁnders, RADAR systems, and color cameras, most of them pointing ahead to detect the road parameters and potential obstacles (e.g. [5], [6], [7]).”
3DmFV: Three-Dimensional Point Cloud Classification in Real-Time Using Convolutional Neural Networks (2018, background): “cameras, providing point clouds, has become a common choice for autonomouse vehicles such as cars [1], and quadrotors [2].”
Continuous 3D scan-matching with a spinning 2D laser (2009, methodology): “Another option is the use of a high-speed multielement 3D lidar such as the Velodyne HDL-64E (e.g., [6], [ 7 ]).”
Behavior Trees in Robotics and AI: An Introduction (2017, methodology): “1 BTs in autonomous vehicles There is no standard control architecture for autonomous vehicles, however reviewing the architectures used to address the DARPA Grand Challenge, a competition for autonomous vehicles, we note that most teams employed FSMs designed and developed exactly for that challeng”
An Open Approach to Autonomous Vehicles (2015, background): “For example, one notable work was published by Carnegie Mellon University.(1) Despite this trend, autonomous vehicles are not systematically organized.”
An Invitation to Imitation (2015, background): “This architecture has been replicated and reﬁned throughout the ﬁeld of robotics [Zucker et al., 2011 , Urmson et al., 2008 , Welling-ton and Stentz, 2004 , Leonard et al., 2008 , Jackel et al., 2006 , Bachrach et al., 2009 ] and is currently used in the most advanced autonomous navigation systems.”
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.”
What could move? Finding cars, pedestrians and bicyclists in 3D laser data (2012, background): “The success of the DARPA Grand- [4] and Urban Challenges [5] as well as Google’s endeavour to promote autonomous driving for data gathering purposes [6] has heightened expectations that autonomous cars will be able to operate in environments of realistic complexity.”
FAST-LIO2: Fast Direct LiDAR-Inertial Odometry (2021, background): “Due to the ability to provide direct, dense, active, and accurate depth measurements of environments, 3D light detection and ranging (LiDAR) sensor has emerged as another essential sensor for robots [9, 10].”