Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Online Learning for Robot Vision
Linköping University, Department of Electrical Engineering, Computer Vision. Linköping University, The Institute of Technology.
2014 (English)Licentiate thesis, comprehensive summary (Other academic)
Abstract [en]

In tele-operated robotics applications, the primary information channel from the robot to its human operator is a video stream. For autonomous robotic systems however, a much larger selection of sensors is employed, although the most relevant information for the operation of the robot is still available in a single video stream. The issue lies in autonomously interpreting the visual data and extracting the relevant information, something humans and animals perform strikingly well. On the other hand, humans have great diculty expressing what they are actually looking for on a low level, suitable for direct implementation on a machine. For instance objects tend to be already detected when the visual information reaches the conscious mind, with almost no clues remaining regarding how the object was identied in the rst place. This became apparent already when Seymour Papert gathered a group of summer workers to solve the computer vision problem 48 years ago [35].

Articial learning systems can overcome this gap between the level of human visual reasoning and low-level machine vision processing. If a human teacher can provide examples of what to be extracted and if the learning system is able to extract the gist of these examples, the gap is bridged. There are however some special demands on a learning system for it to perform successfully in a visual context. First, low level visual input is often of high dimensionality such that the learning system needs to handle large inputs. Second, visual information is often ambiguous such that the learning system needs to be able to handle multi modal outputs, i.e. multiple hypotheses. Typically, the relations to be learned  are non-linear and there is an advantage if data can be processed at video rate, even after presenting many examples to the learning system. In general, there seems to be a lack of such methods.

This thesis presents systems for learning perception-action mappings for robotic systems with visual input. A range of problems are discussed, such as vision based autonomous driving, inverse kinematics of a robotic manipulator and controlling a dynamical system. Operational systems demonstrating solutions to these problems are presented. Two dierent approaches for providing training data are explored, learning from demonstration (supervised learning) and explorative learning (self-supervised learning). A novel learning method fullling the stated demands is presented. The method, qHebb, is based on associative Hebbian learning on data in channel representation. Properties of the method are demonstrated on a vision-based autonomously driving vehicle, where the system learns to directly map low-level image features to control signals. After an initial training period, the system seamlessly continues autonomously. In a quantitative evaluation, the proposed online learning method performed comparably with state of the art batch learning methods.

Place, publisher, year, edition, pages
Linköping: Linköping University Electronic Press, 2014. , 62 p.
Series
Linköping Studies in Science and Technology. Thesis, ISSN 0280-7971 ; 1678
National Category
Computer Vision and Robotics (Autonomous Systems)
Identifiers
URN: urn:nbn:se:liu:diva-110892DOI: 10.3384/lic.diva-110892ISBN: 978-91-7519-228-4 (print)OAI: oai:DiVA.org:liu-110892DiVA: diva2:750053
Presentation
2014-10-24, Visionen, Hus B, Campus Valla, Linköpings universitet, Linköping, 13:15 (Swedish)
Opponent
Supervisors
Funder
EU, FP7, Seventh Framework Programme, 247947Swedish Research Council
Available from: 2014-09-26 Created: 2014-09-26 Last updated: 2016-05-04Bibliographically approved
List of papers
1. Autonomous Navigation and Sign Detector Learning
Open this publication in new window or tab >>Autonomous Navigation and Sign Detector Learning
Show others...
2013 (English)In: IEEE Workshop on Robot Vision(WORV) 2013, IEEE , 2013, 144-151 p.Conference paper, Published paper (Refereed)
Abstract [en]

This paper presents an autonomous robotic system that incorporates novel Computer Vision, Machine Learning and Data Mining algorithms in order to learn to navigate and discover important visual entities. This is achieved within a Learning from Demonstration (LfD) framework, where policies are derived from example state-to-action mappings. For autonomous navigation, a mapping is learnt from holistic image features (GIST) onto control parameters using Random Forest regression. Additionally, visual entities (road signs e.g. STOP sign) that are strongly associated to autonomously discovered modes of action (e.g. stopping behaviour) are discovered through a novel Percept-Action Mining methodology. The resulting sign detector is learnt without any supervision (no image labeling or bounding box annotations are used). The complete system is demonstrated on a fully autonomous robotic platform, featuring a single camera mounted on a standard remote control car. The robot carries a PC laptop, that performs all the processing on board and in real-time.

Place, publisher, year, edition, pages
IEEE, 2013
National Category
Engineering and Technology
Identifiers
urn:nbn:se:liu:diva-86214 (URN)10.1109/WORV.2013.6521929 (DOI)978-1-4673-5647-3 (ISBN)978-1-4673-5646-6 (ISBN)
Conference
IEEE Workshop on Robot Vision (WORV 2013), 15-17 January 2013, Clearwater Beach, FL, USA
Projects
ELLIITETTCUASUK EPSRC: EP/H023135/1
Available from: 2012-12-11 Created: 2012-12-11 Last updated: 2016-06-14
2. Online Learning of Vision-Based Robot Control during Autonomous Operation
Open this publication in new window or tab >>Online Learning of Vision-Based Robot Control during Autonomous Operation
2015 (English)In: New Development in Robot Vision / [ed] Yu Sun, Aman Behal and Chi-Kit Ronald Chung, Springer Berlin/Heidelberg, 2015, 137-156 p.Chapter in book (Refereed)
Abstract [en]

Online learning of vision-based robot control requires appropriate activation strategies during operation. In this chapter we present such a learning approach with applications to two areas of vision-based robot control. In the first setting, selfevaluation is possible for the learning system and the system autonomously switches to learning mode for producing the necessary training data by exploration. The other application is in a setting where external information is required for determining the correctness of an action. Therefore, an operator provides training data when required, leading to an automatic mode switch to online learning from demonstration. In experiments for the first setting, the system is able to autonomously learn the inverse kinematics of a robotic arm. We propose improvements producing more informative training data compared to random exploration. This reduces training time and limits learning to regions where the learnt mapping is used. The learnt region is extended autonomously on demand. In experiments for the second setting, we present an autonomous driving system learning a mapping from visual input to control signals, which is trained by manually steering the robot. After the initial training period, the system seamlessly continues autonomously. Manual control can be taken back at any time for providing additional training.

Place, publisher, year, edition, pages
Springer Berlin/Heidelberg, 2015
Series
Cognitive Systems Monographs, ISSN 1867-4925 ; Vol. 23
National Category
Computer Vision and Robotics (Autonomous Systems)
Identifiers
urn:nbn:se:liu:diva-110891 (URN)10.1007/978-3-662-43859-6_8 (DOI)978-3-662-43858-9 (ISBN)978-3-662-43859-6 (ISBN)
Available from: 2014-09-26 Created: 2014-09-26 Last updated: 2016-06-14Bibliographically approved
3. Biologically Inspired Online Learning of Visual Autonomous Driving
Open this publication in new window or tab >>Biologically Inspired Online Learning of Visual Autonomous Driving
2014 (English)In: Proceedings British Machine Vision Conference 2014 / [ed] Michel Valstar; Andrew French; Tony Pridmore, BMVA Press , 2014, 137-156 p.Conference paper, Published paper (Refereed)
Abstract [en]

While autonomously driving systems accumulate more and more sensors as well as highly specialized visual features and engineered solutions, the human visual system provides evidence that visual input and simple low level image features are sufficient for successful driving. In this paper we propose extensions (non-linear update and coherence weighting) to one of the simplest biologically inspired learning schemes (Hebbian learning). We show that this is sufficient for online learning of visual autonomous driving, where the system learns to directly map low level image features to control signals. After the initial training period, the system seamlessly continues autonomously. This extended Hebbian algorithm, qHebb, has constant bounds on time and memory complexity for training and evaluation, independent of the number of training samples presented to the system. Further, the proposed algorithm compares favorably to state of the art engineered batch learning algorithms.

Place, publisher, year, edition, pages
BMVA Press, 2014
National Category
Computer Vision and Robotics (Autonomous Systems)
Identifiers
urn:nbn:se:liu:diva-110890 (URN)10.5244/C.28.94 (DOI)1901725529 (ISBN)
Conference
British Machine Vision Conference 2014, Nottingham, UK September 1-5 2014
Note

The video contains the online learning autonomous driving system in operation. Data from the system has been synchronized with the video and is shown overlaid. The actuated steering singnal is visualized as the position of a blue dot. The steering signal predicted by the system is visualized by a green circle. During autonomous operation, these two coincide. When the vehicle is controlled manually (training), the word MANUAL is displayed in the video.The first sequence evaluates the ability of the system to stay on the road during road reconfiguration. The results of the first sequence indicate that the system primarily reacts to features on the road, not features in the surrounding area. The second sequence evaluates the multi-modal abilities of the system. After initial training, the vehicle follows the outer track, going straight in the two three-way junctions. By forcing the vehicle to turn right at one intersection, by means of a short application of manual control, a new mode is introduced. When the system later reaches the same intersection, the vehicle either turns or continues straight ahead depending on which of the two modes is the strongest. The ordering of the modes depends on slight variation in the approach to the junction and on noise.The third sequence is longer, evaluating both multi-modal abilities and effects of track reconfiguration. Container: MP4Codec: h264 1280x720

Available from: 2014-09-26 Created: 2014-09-26 Last updated: 2016-06-14Bibliographically approved
4. Combining Vision, Machine Learning and Automatic Control to Play the Labyrinth Game
Open this publication in new window or tab >>Combining Vision, Machine Learning and Automatic Control to Play the Labyrinth Game
2012 (English)In: Proceedings of SSBA, Swedish Symposium on Image Analysis, 2012, 2012Conference paper, Published paper (Other academic)
Abstract [en]

The labyrinth game is a simple yet challenging platform, not only for humans but also for control algorithms and systems. The game is easy to understand but still very hard to master. From a system point of view, the ball behavior is in general easy to model but close to the obstacles there are severe non-linearities. Additionally, the far from flat surface on which the ball rolls provides for changing dynamics depending on the ball position.

The general dynamics of the system can easily be handled by traditional automatic control methods. Taking the obstacles and uneven surface into account would require very detailed models of the system. A simple deterministic control algorithm is combined with a learning control method. The simple control method provides initial training data. As thelearning method is trained, the system can learn from the results of its own actions and the performance improves well beyond the performance of the initial controller.

A vision system and image analysis is used to estimate the ball position while a combination of a PID controller and a learning controller based on LWPR is used to learn to steer the ball through the maze.

National Category
Computer Vision and Robotics (Autonomous Systems)
Identifiers
urn:nbn:se:liu:diva-110888 (URN)
Conference
Swedish Symposium on Image Analysis for 2012, March 8-9, Stockholm, Sweden
Available from: 2014-09-26 Created: 2014-09-26 Last updated: 2016-06-14Bibliographically approved

Open Access in DiVA

Online Learning for Robot Vision(6420 kB)783 downloads
File information
File name FULLTEXT01.pdfFile size 6420 kBChecksum SHA-512
a96235f8e097cda1c8005438aaa290099ae3b279772e5abe9113d4992aed54fcb21982bd5284a5b2017dd77fedfde3add08abb86744c09f5005242420e6bdeb5
Type fulltextMimetype application/pdf
omslag(7590 kB)29 downloads
File information
File name COVER01.pdfFile size 7590 kBChecksum SHA-512
22f33eb439dbf0e01e62a350dd91b42aac85e1e92313348eba4b54f77567b10c5cdcae6376531dbfc71c12fe93f964a991a556e1248900cd162d9734cb5a923f
Type coverMimetype application/pdf

Other links

Publisher's full text

Search in DiVA

By author/editor
Öfjäll, Kristoffer
By organisation
Computer VisionThe Institute of Technology
Computer Vision and Robotics (Autonomous Systems)

Search outside of DiVA

GoogleGoogle Scholar
Total: 783 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

doi
isbn
urn-nbn

Altmetric score

doi
isbn
urn-nbn
Total: 1364 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf