Detecting and modeling doors with mobile robots

  title={Detecting and modeling doors with mobile robots},
  author={Dragomir Anguelov and Daphne Koller and Evan Parker and Sebastian Thrun},
  journal={IEEE International Conference on Robotics and Automation, 2004. Proceedings. ICRA '04. 2004},
  pages={3777-3784 Vol.4}
  • Dragomir AnguelovD. Koller S. Thrun
  • Published 6 July 2004
  • Computer Science
  • IEEE International Conference on Robotics and Automation, 2004. Proceedings. ICRA '04. 2004
We describe a probabilistic framework for detection and modeling of doors from sensor data acquired in corridor environments with mobile robots. The framework captures shape, color, and motion properties of door and wall objects. The probabilistic model is optimized with a version of the expectation maximization algorithm, which segments the environment into door and wall objects and learns their properties. The framework allows the robot to generalize the properties of detected object… 

Figures from this paper

Probabilistic Mobile Manipulation in Dynamic Environments, with Application to Opening Doors

This paper proposes a unified approach to these two problems that dynamically models the objects to be manipulated and localizes the robot at the same time, and applies this approach to the task of navigating from one office to another (including manipulating doors).

Improving Door Detection for Mobile Robots by fusing Camera and Laser-Based Sensor Data

This paper presents a set of simple door detection classifiers, which are based either on camera or laser-based data and accomplish only a weak door detection rate, but by combining them through a AdaBoost Algorithm more than 82% of all doors with a false positive rate less than 3% are detected in static test data.

Automated Door Detection with a 3D-Sensor

A novel door detection algorithm is proposed that employs basic structural knowledge about doors and enables to extract parts of doors from point clouds based on constraint region growing and gets weighted with Gaussian probabilities and combined to create an overall probability measure.

Visual door detection integrating appearance and shape cues

AdaBoost based Door Detection for Mobile Robots

A cameraand laser-based approach which allows finding more than 82% of all doors with a false positive rate less than 3% in static test sets and by using different door perspectives from a moving robot detects more than 90% ofdoors with a very low false detection rate.

A Probabilistic Framework for Learning Kinematic Models of Articulated Objects

This work presents a novel, probabilistic framework for modeling articulated objects as kinematic graphs, and demonstrates that this approach has a broad set of applications, in particular for the emerging fields of mobile manipulation and service robotics.

Visual detection of lintel-occluded doors from a single image

  • Zhichao ChenStan Birchfield
  • Computer Science
    2008 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops
  • 2008
This work presents a vision-based door detection algorithm that achieves robustness by utilizing a variety of features, including color, texture, and intensity edges, and introduces two novel geometric features that increase performance significantly: concavity and bottom-edge intensity profile.

Laser-based perception for door and handle identification

The approach builds on a 3D perception pipeline to annotate doors and their handles solely from sensed laser data, without any a priori model learning, to segment the parts of interest using robust geometric estimators and statistical methods applied on geometric and intensity distribution variations in the scan.

Learning to open new doors

This paper proposes an approach that, rather than trying to build a full 3d model of the door/door handle, uses computer vision to choose a manipulation strategy, and evaluates on a large set of doors that the robot had not previously seen, it successfully opened 31 out of 34 doors.

Simple monocular door detection and tracking

A new framework for door detection and tracking which exploits geometrical features of corridors and relies on visual features such as lines and vanishing points that are combined to discriminate the floor and wall planes and then to recognize doors within the image sequences.



Learning Hierarchical Object Maps of Non-Stationary Environments with Mobile Robots

This paper presents an algorithm for learning object models of non-stationary objects found in office-type environments through a two-level hierarchical representation that outperforms a previously developed non-hierarchical algorithm that models objects but lacks class templates.

Map building with mobile robots in populated environments

This paper uses a probabilistic method to track multiple people and to incorporate the results of the tracking technique into the mapping process, which results in more accurate maps.

A Distributed Model for Mobile Robot Environment-Learning and Navigation

A distributed method for mobile robot navigation, spatial learning, and path planning is presented and the main issues addressed are: distributed, procedural, and qualitative representation and computation, emergent behaviors, dynamic landmarks, minimized communication.

Using EM to Learn 3D Models of Indoor Environments with Mobile Robots

An algorithm for generating compact 3D models of indoor environments with mobile robots using the expectation maximization algorithm to fit a lowcomplexity planar model to 3D data collected by range finders and a panoramic camera is described.

Position referencing and consistent world modeling for mobile robots

  • R. ChatilaJ. Laumond
  • Computer Science
    Proceedings. 1985 IEEE International Conference on Robotics and Automation
  • 1985
The approach proposed in this paper relies on the use of a multisensory system, favo ring of the data collected by the more accurate sensor in a given situation, averaging of different but consistent measurements of the same entity weighted with their associated uncertainties.

A robot exploration and mapping strategy based on a semantic hierarchy of spatial representations

MAGELLAN: An Integrated Adaptive Architecture for Mobile Robotics

The system represents its spatial knowledge in terms of a topological network that connects a set of distinct places, each represented by evidence grids that contain probabilistic descriptions of occupancy, suggesting that the system can operate robustly across a range of environments, including ones that involve dynamic changes.

An Atlas framework for scalable mapping

Atlas is described, a hybrid metrical/topological approach to SLAM that achieves efficient mapping of large-scale environments using a graph of coordinate frames that captures the local environment and the current robot pose along with the uncertainties of each.

A real-time algorithm for mobile robot mapping with applications to multi-robot and 3D mapping

  • S. ThrunW. BurgardD. Fox
  • Computer Science
    Proceedings 2000 ICRA. Millennium Conference. IEEE International Conference on Robotics and Automation. Symposia Proceedings (Cat. No.00CH37065)
  • 2000
We present an incremental method for concurrent mapping and localization for mobile robots equipped with 2D laser range finders. The approach uses a fast implementation of scan-matching for mapping,

Incremental mapping of large cyclic environments

  • Jens-Steffen GutmannK. Konolige
  • Computer Science
    Proceedings 1999 IEEE International Symposium on Computational Intelligence in Robotics and Automation. CIRA'99 (Cat. No.99EX375)
  • 1999
A method, called local registration and global correlation, for reliable reconstruction of consistent global maps from dense range data, is presented, attractive because it is incremental, producing an updated map with every new sensor input; and runs in constant time independent of the size of the map.