Off-Road Path and Obstacle Detection Using Decision Networks and Stereo Vision

Caraffi, C.; Cattani, Silvia; Grisleri, Paolo

doi:10.1109/tits.2007.908583

Cited by 70 publications

(34 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Generally, they are classified as four methods which focus on different aspects: disparity property-based method, occupancy grid-based method, geometry-based method, and learningbased method. V-disparity 6 is a prime approach in free space detection using disparity property and many other methods are based on this idea 5,7,8 It takes the disparity map as input and accumulates the number of pixels with the same disparity values along each row, obtaining a new map whose rows are the image rows and columns are the disparities sorted increasingly. Therefore, each row in this v-disparity map often contains a dominant value which shows that most of the pixels in this row take the same disparity.…”

Section: Related Workmentioning

confidence: 99%

Gaussian process regression-based robust free space detection for autonomous vehicle by 3-D point cloud and 2-D appearance information fusion

Xiao

Dai

et al. 2017

International Journal of Advanced Robotic Systems

View full text Add to dashboard Cite

Free space detection is crucial to autonomous vehicles while existing works are not entirely satisfactory. As cameras have many advantages on environment perception, a stereo vision-based robust free space detection method is proposed which mainly depends on geometry information and Gaussian process regression. In this work, in order to improve the performance by exploiting multiple source information, we apply Bayesian framework and conditional random field inference to fuse the multimodal information including 2-D image and 3-D point geometric information. Particularly, a Bayesian framework is used for multiple feature fusion to provide a normalized and flexible output. Gaussian process regression is used to automatically and incrementally regress the data, resulting enhanced performance. Finally, conditional random field with color and geometry constrains is applied to make the result more robust. In order to evaluate the proposed method, quantitative experiments on popular KITTI-road data set and qualitative experiments on our own campus data set are tested. The results show satisfactory and inspiring performance compared to the outstanding works and even are competitive to some relevant Lidar-based methods.

show abstract

Section: Related Workmentioning

confidence: 99%

Gaussian process regression-based robust free space detection for autonomous vehicle by 3-D point cloud and 2-D appearance information fusion

Xiao

Dai

et al. 2017

International Journal of Advanced Robotic Systems

View full text Add to dashboard Cite

show abstract

“…& Vehicle positioning (GPS receiver) and kinematical variables measurement (vehicle CAN bus) & Obstacles detection subsystem that can be based on radar, laser-scanner or computer vision, among other technologies, or a combination of some of them using Vehicle 1 hits front part Vehicle 2 TD13 > TD23 TA21 > TD11 G I TA21 > TD23 * TD13 > TD11 * (max(TD11,TD23) *0Always occurs X0No accident sensor fusion techniques [12][13][14]. This subsystem involves also obstacles detection and tracking algorithms [10].…”

Section: Collision Avoidance Systemmentioning

confidence: 99%

An Improved Method to Calculate the Time-to-Collision of Two Vehicles

2012

View full text Add to dashboard Cite

In order to improve vehicle safety, a interaction phase between primary and secondary safety systems has been defined. These systems use information provided by the primary safety systems to achieve both the primary and the secondary systems' objectives. It is essential to discriminate whether a collision is avoidable or not and to calculate the time available before the crash happens. This paper shows a method that improves on other simplified methods to calculate the time-to-collision (TTC) to provide a more accurate result that could be used in a collision avoidance system.

show abstract

“…Driving in urban traffic requires detailed perception of the environment surrounding the vehicle: for this, we installed in the truck cabin a trinocular vision system capable of performing both obstacle and lane detection up to distances of 40 m, derived from Caraffi, Cattani, and Grisleri (2007). The stereo approach has been chosen because it allows an accurate three-dimensional (3D) reconstruction without requiring strong a priori knowledge of the scene in front of the vehicle, but just correct calibration values, which are being estimated at run time.…”

Section: The Trinocular Systemmentioning

confidence: 99%

TerraMax™: Team Oshkosh urban robot

Chen

Sundareswaran

Craig

et al. 2008

Journal of Field Robotics

View full text Add to dashboard Cite

University, participated in the DARPA Urban Challenge and was one of the 11 teams selected to compete in the final event. Through development, testing, and participation in the official events, we experimented and demonstrated autonomous truck operations in (controlled) urban streets of California, Wisconsin, and Michigan under various climate and traffic conditions. In these experiments TerraMax TM , a modified medium tactical vehicle replacement (MTVR) truck by Oshkosh Corporation, negotiated urban roads, intersections, and parking lots and interacted with manned and unmanned traffic while observing traffic rules. We accumulated valuable experience and lessons on autonomous truck operations in urban environments, particularly in the aspects of vehicle control, perception, mission planning, and autonomous behaviors, which will have an impact on the further development of large-footprint autonomous ground vehicles for the military. • Journal of Field Robotics-2008In this article, we describe the vehicle, the overall system architecture, the sensors and sensor processing, the mission planning system, and the autonomous behavioral controls implemented on TerraMax. We discuss the performance of some notable autonomous behaviors of TerraMax and our experience in implementing these behaviors and present results of the Urban Challenge National Qualification Event tests and the Urban Challenge Final Event. We conclude with a discussion of lessons learned from all of the above experience in working with a large robotic truck. C 2008 Wiley Periodicals, Inc.

show abstract

Off-Road Path and Obstacle Detection Using Decision Networks and Stereo Vision

Cited by 70 publications

References 20 publications

Gaussian process regression-based robust free space detection for autonomous vehicle by 3-D point cloud and 2-D appearance information fusion

Gaussian process regression-based robust free space detection for autonomous vehicle by 3-D point cloud and 2-D appearance information fusion

An Improved Method to Calculate the Time-to-Collision of Two Vehicles

TerraMax™: Team Oshkosh urban robot

Contact Info

Product

Resources

About