Mobile robot self-localization in complex indoor environments using monocular vision and 3D model

Kitanov, Andrej; Bisevac, S.; Petrović, Ivan

doi:10.1109/aim.2007.4412566

Cited by 22 publications

(12 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The GPU has been frequently used in robot localization for precisely this reason, including: Kinect depth-SLAM [17], image feature correspondence search for SIFT features [18], and line features [19].…”

Section: Related Workmentioning

confidence: 99%

Visual localization within LIDAR maps for automated urban driving

Wolcott

Eustice

2014

2014 IEEE/RSJ International Conference on Intelligent Robots and Systems

329

147

View full text Add to dashboard Cite

Abstract-This paper reports on the problem of map-based visual localization in urban environments for autonomous vehicles. Self-driving cars have become a reality on roadways and are going to be a consumer product in the near future. One of the most significant road-blocks to autonomous vehicles is the prohibitive cost of the sensor suites necessary for localization. The most common sensor on these platforms, a three-dimensional (3D) light detection and ranging (LIDAR) scanner, generates dense point clouds with measures of surface reflectivity-which other state-of-the-art localization methods have shown are capable of centimeter-level accuracy. Alternatively, we seek to obtain comparable localization accuracy with significantly cheaper, commodity cameras. We propose to localize a single monocular camera within a 3D prior groundmap, generated by a survey vehicle equipped with 3D LIDAR scanners. To do so, we exploit a graphics processing unit to generate several synthetic views of our belief environment. We then seek to maximize the normalized mutual information between our real camera measurements and these synthetic views. Results are shown for two different datasets, a 3.0 km and a 1.5 km trajectory, where we also compare against the state-of-the-art in LIDAR map-based localization.

show abstract

Section: Related Workmentioning

confidence: 99%

Visual localization within LIDAR maps for automated urban driving

Wolcott

Eustice

2014

2014 IEEE/RSJ International Conference on Intelligent Robots and Systems

329

147

View full text Add to dashboard Cite

show abstract

“…In [4] a vision-based method is described that uses a 3D model of the environment to localize itself. This allows for a nearly unlimited size environment and does not require a specific part of it to be in view.…”

Section: Related Workmentioning

confidence: 99%

“…Many vision-based control solutions for MAVs and other robots pose significant processing power demands [1], [2], [4], [8], mainly due to the computational costs associated with image processing and computer vision algorithms. It has been shown that the PixArt IR tracking sensor can be successfully applied in a visual servoing algorithm on a quadrotor vehicle [3], without the need for powerful processing hardware.…”

Section: Related Workmentioning

confidence: 99%

Low-cost vision-based 6-DOF MAV localization using IR beacons

Roozing

Göktoğan

2013

2013 IEEE/ASME International Conference on Advanced Intelligent Mechatronics

View full text Add to dashboard Cite

The autonomous operation of Micro Aerial Vehicles (MAVs) is a challenging area of research that has received much research interest in recent years. Particularly, accurate localisation of MAVs is an issue, most important during tasks that require high accuracy, such as indoor flight and landing. Many existing solutions are either not accurate, heavy or expensive. We present a low-cost vision-based solution to the localisation problem of MAVs. An on-board infrared tracking sensor with built-in vision processing is used to detect infrared markers and a point-based pose estimation algorithm is implemented to obtain 6 DOF localisation at high rates. The system performance is compared against Inertial Measurement Unit (IMU) and external stereo vision measurements as ground truth. We show that the system produces accurate 6 DOF estimates at low cost and weight in a computationally inexpensive way, and are immediately usable in control implementations. Thus our solution provides a feasible localization solution for MAVs.

show abstract

“…Kitanov et al [13] and Koch and Teller [14] present near neighbors to our approach, using hand-built 3D models of an environment. Images rendered from this model then have features extracted; these features are compared to features extracted from camera images of the actual scene.…”

Section: Prior Workmentioning

confidence: 99%

Textured occupancy grids for monocular localization without features

Mason

Ricco

Parr

2011

2011 IEEE International Conference on Robotics and Automation

View full text Add to dashboard Cite

Abstract-A textured occupancy grid map is an extremely versatile data structure. It can be used to render humanreadable views and for laser rangefinder localization algorithms. For camera-based localization, landmark or featurebased maps tend to be favored in current research. This may be because of a tacit assumption that working with a textured occupancy grid with a camera would be impractical. We demonstrate that a textured occupancy grid can be combined with an extremely simple monocular localization algorithm to produce a viable localization solution. Our approach is simple, efficient, and produces localization results comparable to laser localization results. A consequence of this result is that a single map representation, the textured occupancy grid, can now be used for humans, robots with laser rangefinders, and robots with just a single camera.

show abstract

Mobile robot self-localization in complex indoor environments using monocular vision and 3D model

Cited by 22 publications

References 7 publications

Visual localization within LIDAR maps for automated urban driving

Visual localization within LIDAR maps for automated urban driving

Low-cost vision-based 6-DOF MAV localization using IR beacons

Textured occupancy grids for monocular localization without features

Contact Info

Product

Resources

About