Results 1 - 10
of
32
Selfsupervised monocular road detection in desert terrain
- In Proc. of Robotics: Science and Systems (RSS
, 2006
"... Abstract — We present a method for identifying drivable surfaces in difficult unpaved and offroad terrain conditions as encountered in the DARPA Grand Challenge robot race. Instead of relying on a static, pre-computed road appearance model, this method adjusts its model to changing environments. It ..."
Abstract
-
Cited by 107 (6 self)
- Add to MetaCart
(Show Context)
Abstract — We present a method for identifying drivable surfaces in difficult unpaved and offroad terrain conditions as encountered in the DARPA Grand Challenge robot race. Instead of relying on a static, pre-computed road appearance model, this method adjusts its model to changing environments. It achieves robustness by combining sensor information from a laser range finder, a pose estimation system and a color camera. Using the first two modalities, the system first identifies a nearby patch of drivable surface. Computer Vision then takes this patch and uses it to construct appearance models to find drivable surface outward into the far range. This information is put into a drivability map for the vehicle path planner. In addition to evaluating the method’s performance using a scoring framework run on real-world data, the system was entered, and won, the 2005 DARPA Grand Challenge. Post-race log-file analysis proved that without the Computer Vision algorithm, the vehicle would not have driven fast enough to win. I.
Learning from Demonstration for Autonomous Navigation in Complex Unstructured Terrain
, 2010
"... Rough terrain autonomous navigation continues to pose a challenge to the robotics community. Robust navigation by a mobile robot depends not only on the individual performance of perception and planning systems, but on how well these systems are coupled. When traversing complex unstructured terrain, ..."
Abstract
-
Cited by 17 (7 self)
- Add to MetaCart
(Show Context)
Rough terrain autonomous navigation continues to pose a challenge to the robotics community. Robust navigation by a mobile robot depends not only on the individual performance of perception and planning systems, but on how well these systems are coupled. When traversing complex unstructured terrain, this coupling (in the form of a cost function) has a large impact on robot behavior and performance, necessitating a robust design. This paper explores the application of Learning from Demonstration to this task for the Crusher autonomous navigation platform. Using expert examples of desired navigation behavior, mappings from both online and offline perceptual data to planning costs are learned. Challenges in adapting existing techniques to complex online planning systems and imperfect demonstration are addressed, along with additional practical considerations. The benefits to autonomous performance of this approach are examined, as well as the decrease in necessary designer effort. Experimental results are presented from autonomous traverses through complex natural environments. 1
Unsupervised learning of invariant features using video
- In CVPR
, 2010
"... We present an algorithm that learns invariant features from real data in an entirely unsupervised fashion. The principal benefit of our method is that it can be applied without human intervention to a particular application or data set, learning the specific invariances necessary for excellent featu ..."
Abstract
-
Cited by 15 (1 self)
- Add to MetaCart
(Show Context)
We present an algorithm that learns invariant features from real data in an entirely unsupervised fashion. The principal benefit of our method is that it can be applied without human intervention to a particular application or data set, learning the specific invariances necessary for excellent feature performance on that data. Our algorithm relies on the ability to track image patches over time using optical flow. With the wide availability of high frame rate video (eg: on the web, from a robot), good tracking is straightforward to achieve. The algorithm then optimizes feature parameters such that patches corresponding to the same physical location have feature descriptors that are as similar as possible while simultaneously maximizing the distinctness of descriptors for different locations. Thus, our method captures data or application specific invariances yet does not require any manual supervision. We apply our algorithm to learn domain-optimized versions of SIFT and HOG. SIFT and HOG features are excellent and widely used. However, they are general and by definition not tailored to a specific domain. Our domain-optimized versions offer a substantial performance increase for classification and correspondence tasks we consider. Furthermore, we show that the features our method learns are near the optimal that would be achieved by directly optimizing the test set performance of a classifier. Finally, we demonstrate that the learning often allows fewer features to be used for some tasks, which has the potential to dramatically improve computational concerns for very large data sets. 1.
Laser-based navigation enhanced with 3d time-of-flight data
- In Proce ICRA ’09
, 2009
"... Abstract — Navigation and obstacle avoidance in robotics using planar laser scans has matured over the last decades. They basically enable robots to penetrate highly dynamic and populated spaces, such as people’s home, and move around smoothly. However, in an unconstrained environment the two-dimens ..."
Abstract
-
Cited by 13 (2 self)
- Add to MetaCart
(Show Context)
Abstract — Navigation and obstacle avoidance in robotics using planar laser scans has matured over the last decades. They basically enable robots to penetrate highly dynamic and populated spaces, such as people’s home, and move around smoothly. However, in an unconstrained environment the two-dimensional perceptual space of a fixed mounted laser is not sufficient to ensure safe navigation. In this paper, we present an approach that pools a fast and reliable motion generation approach with modern 3D capturing techniques using a Time-of-Flight camera. Instead of attempting to implement full 3D motion control, which is computationally more expensive and simply not needed for the targeted scenario of a domestic robot, we introduce a “virtual laser”. For the originally solely laser-based motion generation the technique of fusing real laser measurements and 3D point clouds into a continuous data stream is 100 % compatible and transparent. The paper covers the general concept, the necessary extrinsic calibration of two very different types of sensors, and exemplarily illustrates the benefit which is to avoid obstacles not being perceivable in the original laser scan. I.
Online Learning for Offroad Robots: Using Spatial Label Propagation to Learn
"... Abstract — We present a solution to the problem of long-range obstacle/path recognition in autonomous robots. The system uses sparse traversability information from a stereo module to train a classifier online. The trained classifier can then predict the traversability of the entire scene. A distanc ..."
Abstract
-
Cited by 10 (5 self)
- Add to MetaCart
(Show Context)
Abstract — We present a solution to the problem of long-range obstacle/path recognition in autonomous robots. The system uses sparse traversability information from a stereo module to train a classifier online. The trained classifier can then predict the traversability of the entire scene. A distance-normalized image pyramid makes it possible to efficiently train on each frame seen by the robot, using large windows that contain contextual information as well as shape, color, and texture. Traversability labels are initially obtained for each target using a stereo module, then propagated to other views of the same target using temporal and spatial concurrences, thus training the classifier to be viewinvariant. A ring buffer simulates short-term memory and ensures that the discriminative learning is balanced and consistent. This long-range obstacle detection system sees obstacles and paths at 30-40 meters, far beyond the maximum stereo range of 12 meters, and adapts very quickly to new environments. Experiments were run on the LAGR robot platform. I.
Utilizing prior information to enhance self-supervised aerial image analysis for extracting parking lot structures
- In Proceedings of IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS-2009
, 2009
"... Abstract — Road network information (RNI) simplifies autonomous driving by providing strong priors about driving environments. Its usefulness has been demonstrated in the DARPA Urban Challenge. However, the need to manually generate RNI prevents us from fully exploiting its benefits. We envision an ..."
Abstract
-
Cited by 6 (2 self)
- Add to MetaCart
(Show Context)
Abstract — Road network information (RNI) simplifies autonomous driving by providing strong priors about driving environments. Its usefulness has been demonstrated in the DARPA Urban Challenge. However, the need to manually generate RNI prevents us from fully exploiting its benefits. We envision an aerial image analysis system that automatically generates RNI for a route between two urban locations. As a step toward this goal, we present an algorithm that extracts the structure of a parking lot visible in an aerial image. We formulate this task as a problem of parking spot detection because extracting parking lot structures is closely related to detecting all of the parking spots. To minimize human intervention in use of aerial imagery, we devise a self-supervised learning algorithm that automatically obtains a set of canonical parking spot templates to learn the appearance of a parking lot and estimates the structure of the parking lot from the learned model. The data set extracted from a single image alone is too small to sufficiently learn an accurate parking spot model. To remedy this insufficient positive data problem, we utilize selfsupervised parking spots obtained from other aerial images as prior information and a regularization technique to avoid an overfitting solution. I.
Experimental Evaluation of State of the Art 3D-Sensors for Mobile Robot Navigation 1)
"... In this paper we discuss the suitability of three optical 3D range finders for mobile robot navigation in indoor environments. The range finders under consideration are the CSEM SwissRanger time-of-flight camera 1) and both, a horizontal and vertical baseline stereo-vision system. Modelling the worl ..."
Abstract
-
Cited by 5 (0 self)
- Add to MetaCart
(Show Context)
In this paper we discuss the suitability of three optical 3D range finders for mobile robot navigation in indoor environments. The range finders under consideration are the CSEM SwissRanger time-of-flight camera 1) and both, a horizontal and vertical baseline stereo-vision system. Modelling the world in 2D and using 2D sensors that scan parallel to the ground for the task of navigation is still popular in the robotics community. This approach works in adapted, but not in general indoor environments. We evaluate the usability of named 3D sensors in such general environments by practical experiments. 1
Augmenting Cartographic Resources for Autonomous Driving
, 2009
"... In this paper we present algorithms for automatically generating a road network description from aerial imagery. The road network inforamtion (RNI) produced by our algorithm includes a composite topoloigical and spatial representation of the roads visible in an aerial image. We generate this data fo ..."
Abstract
-
Cited by 5 (4 self)
- Add to MetaCart
In this paper we present algorithms for automatically generating a road network description from aerial imagery. The road network inforamtion (RNI) produced by our algorithm includes a composite topoloigical and spatial representation of the roads visible in an aerial image. We generate this data for use by autonomous vehicles operating on-road in urban environments. This information is used by the vehicles to both route plan and determine appropriate tactical behaviors. RNI can provide important contextual cues that influence driving behaviors, such as the curvature of the road ahead, the location of traffic signals, or pedestrian dense areas. The value of RNI was demonstrated compellingly in the DARPA Urban Challenge 1, where the vehicles relied on this information to drive quickly, safely and efficiently. The current best methods for generating RNI are manual, labor intensive and error prone. Automation of this process could thus provide an important capability. As a step toward this goal, we present algorithms that automatically build the skeleton of drivable regions in a parking lot from a single orthoimage. As a first step in extracting structure, our algorithm detects the parking spots visible in an image. It then combines this information with the detected parking lot boundary and information from other detected
A multi-range architecture for collision-free off-road robot navigation
- Journal of Field Robotics
"... We present a multi-layered mapping, planning, and command execution system developed and tested on the LAGR mobile robot. Key to robust performance under uncertainty is the combination of a short-range perception system operating at high frame rate and low resolution and a long-range, adaptive visio ..."
Abstract
-
Cited by 5 (1 self)
- Add to MetaCart
(Show Context)
We present a multi-layered mapping, planning, and command execution system developed and tested on the LAGR mobile robot. Key to robust performance under uncertainty is the combination of a short-range perception system operating at high frame rate and low resolution and a long-range, adaptive vision system operating at lower frame rate and higher resolution. The short-range module performs local planning and obstacle avoidance with fast reaction times, while the long-range module performs strategic visual planning. Probabilistic traversability labels provided by the perception modules are combined and accumulated into a robot-centered hyperbolic-polar map with a 200 meter effective range. Instead of using a dynamical model of the robot for short-range planning, the system uses a large lookup table of physically-possible trajectory segments recorded on the robot in a wide variety of driving conditions. Localization is performed using a combination of GPS, wheel odometry, IMU, and a high-speed, low-complexity rotational visual odometry module. The end to end system was developed and tested on the LAGR mobile robot, and was verified in independent government tests. 1
Self-Supervised Aerial Image Analysis for Extracting Parking Lot Structure
"... Road network information simplifies autonomous driving by providing strong priors about environments. It informs a robotic vehicle with where it can drive, models of what can be expected, and contextual cues that influence driving behaviors. Currently, however, road network information is manually g ..."
Abstract
-
Cited by 4 (2 self)
- Add to MetaCart
Road network information simplifies autonomous driving by providing strong priors about environments. It informs a robotic vehicle with where it can drive, models of what can be expected, and contextual cues that influence driving behaviors. Currently, however, road network information is manually generated using a combination of GPS survey and aerial imagery. These manual techniques are labor intensive and error prone. To fully exploit the benefits of digital imagery, these processes should be automated. As a step toward this goal, we present an algorithm that extracts the structure of a parking lot visible from a given aerial image. To minimize human intervention in the use of aerial imagery, we devise a self-supervised learning algorithm that automatically generates a set of parking spot templates to learn the appearance of a parking lot and estimates the structure of the parking lot from the learned model. The data set extracted from a single image alone is too small to sufficiently learn an accurate parking spot model. However, strong priors trained using large data sets collected across multiple images dramatically improve performance. Our self-supervised approach outperforms the prior alone by adapting the distribution of examples toward that found in the current image. A thorough empirical analysis compares leading state-of-the-art learning techniques on this problem.
