Vision-based door-traversal for autonomous mobile robots

Eberst, Christof; Andersson, Magnus; Christensen, Henrik I.

doi:10.1109/iros.2000.894673

Cited by 18 publications

(13 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Muñoz-Salinas et al (2005) present a visual door detection system that is based on the Canny edge detector and Hough transform to extract line segments from images. Another vision-based system for detection and traversal of doors is presented in (Eberset et al, 2000). Door structures are extracted from images using a line based filtering parallel method, and an active tracking of detected door line segments is used to drive the robot through the door.…”

Section: The Door Identification Taskmentioning

confidence: 99%

Approaches to Door Identification for Robot Navigation

Jauregi¹,

Lazkano²,

Sierra³

2010

Mobile Robots Navigation

View full text Add to dashboard Cite

Section: The Door Identification Taskmentioning

confidence: 99%

Approaches to Door Identification for Robot Navigation

Jauregi¹,

Lazkano²,

Sierra³

2010

Mobile Robots Navigation

View full text Add to dashboard Cite

“…In Ref. 15, a vision-based door-traversal behavior is presented. Door recognition is based on line segments extraction and is carried out through a matching between a predicted model of the door-frame and the extracted parallel line segments opportunely filtered.…”

Section: Recognizing the Doormentioning

confidence: 99%

Different Learning Methodologies for Vision-Based Navigation Behaviors

Cicirelli

D’Orazio

Distante

2005

Int. J. Patt. Recogn. Artif. Intell.

View full text Add to dashboard Cite

In this work the complex behavior of localizing a mobile vehicle with respect to the door of the environment and then reaching the door has been developed. The robot uses visual information to detect and recognize the door and to determine its state with respect to it. This complex task has been divided into two separate behaviors: door-recognition and door-reaching. A supervised methodology based on learning by components has been applied for recognizing the door. Learning by components allows to recognize the door also in difficult situations such as partial occlusions and besides, it makes recognition independent of viewpoint variations and scale changes. An unsupervised methodology based on reinforcement learning has been used for the door-reaching behavior, instead. The image of the door gives information about the relative position of the vehicle with respect to the door. Then the Q-learning algorithm is used to generate the optimal state-action associations. The problem of defining the state and the action sets has been addressed with the aim of producing smooth paths, of reducing the effects of visual errors during real navigation, and of keeping low the computational cost during the learning phase. A novel way to obtain a continuous action set has been introduced: it uses a fuzzy model to evaluate the system state. Experimental results in real environment show both the robustness of the door-recognition behavior and the generality of the door-reaching behavior.

show abstract

“…On the other hand, a vision-based system for detection and traversal of doors is presented in [6]. Door structures are extracted from images using a parallel line based filtering method, and an active tracking of detected door line segments is used to drive the robot through the door.…”

Section: Introductionmentioning

confidence: 99%

Tartalo: the door knocker robot

Jauregi¹,

Martínez-Otzeta²,

Sierra³

et al. 2007

2007 IEEE International Conference on Robotics and Biomimetics (ROBIO)

View full text Add to dashboard Cite

Bioinspired navigation strategies require landmark identification subsystems for goal identification. Door recognition is a key problem to be solved during mobile robot navigation. Doors give access to many locations that are defined as goals for the robot. This paper presents an approach to door identification by means of recognition of the door handle. Rather than using the lines defined by the door blades, the region of interest of an image is extracted by means of the Hough transform and afterwards SIFT keypoints are obtained and matched against a database in order to positively recognize the door. The extraction of the ROI highly reduces the computational load of the SIFT algorithm and increases the handle recognition performance. The approach is evaluated and tested on a real Peoplebot robot, using a behavior-based control architecture that combines the identification module with free-space balancing (for corridor following), door knocking and crossing behaviors.Index Terms-Scale invariant feature transform, behaviorbased robotics

show abstract

Vision-based door-traversal for autonomous mobile robots

Cited by 18 publications

References 10 publications

Approaches to Door Identification for Robot Navigation

Approaches to Door Identification for Robot Navigation

Different Learning Methodologies for Vision-Based Navigation Behaviors

Tartalo: the door knocker robot

Contact Info

Product

Resources

About