Control Systems and Computers, N6, 2018, Article 3

DOI: https://doi.org/10.15407/usim.2018.06.036

Upr. sist. maš., 2018, Issue 6 (278), pp. 36-45.

UDC 004.896

Sukhoruchkina Olga N., Senior researches, E-mail: sukhoru@irtc.org.ua,

Progonnyi Nikolay V., Researches, E‑mail: progonny@gmail.com,

International Research and Training Center for Informational Technologies and Systems of the National Academy of Sciences of Ukraine and Ministry of Education and Science of Ukraine, 40 Academician Glushkov av., Kiev 03187 Ukraine

On the formation of the object models by autonomous mobile robot resources

Introduction. A modern mobile robot (MR) of a service type is a technical system that autonomously moves and performs various missions of the user in a dynamical changing environment. For safe and efficient of the MR functioning, first of all it is necessary to identify the spatial characteristics of the surrounding objects as the obstacles to robot movement. However, if the MR mission is related to the search, pursuit, movement of certain objects or the approach to them, the information technology (IT) of the automatic formation of the objects models as the long-term MR memory specialized structures is topical for future recognitions this objects in environment.

Purpose — representation of the IT of the automatic formation of the object models, based on the sensor-range finder data and an on-board camera image of the MR ERIC (our MR experimental sample). These models should contain both spatial characteristics and visual features of the objects.

Methods. The presented IT based on the use of methods for analyzing and interpreting sets of range finders measurements for constructing a geometric object model and on the algorithms for the finding and description of key features of an object image. To do this, we propose a technique for the automatic synthesis of a reference image based on the current on-board camera images. The reference image contains only the target object without any background elements from the environment.

Results. The proposed IT is used as the basic information functionality for the algorithm of autonomous MR mission “Examine object”. When the MR ERIC autonomously moves around the target object, the special structures of its long-term memory are filled with the spatial and visual characteristics of this object.

Conclusion. The proposed IT allows you to abandon the stage of forming a library of visual features of objects with the participation of a person. The sequential analysis of two type of information about the target object — from range finder and images from MR on-board camera, which corresponds to the same position of the MR in relation to the object, has allowed to automatically synthesize the reference images and formed the model of this object as a set of its spatial parameters and visual features. Formed models can be used in the autonomous execution of MR missions for which objects with known models are the target objects but their positions in space are initially unknown.

 Download full text! (In Russian).

Keywords: autonomous mobile robot, sensor subsystem, spatial perception, object model.

  1. Sukhoruchkina, O.N., 2007. “The structures of functional organization of the mobile cybernetic system intelligent control”. Upravlausie sistemy i masiny, 3, pp. 26–33, 63. (In Russian).
  2. Sukhoruchkina, O.N., 2012. “The structures and information processes of the mobile robot intelligent control”. Zbіrnyk naukovykh prats Instytutu problem modelyuvannya v energetytsi іm. Ye. Pukhova NAN Ukrainy, Kyiv, 62. pp. 93–101. (In Russian).
  3. Controller Area Network (CAN), an overview, [online] Available at: <http://www.can-cia.de/can/> [Accessed 16 Jan. 2018].
  4. Sukhoruchkina, O.N., Progonnyi, N.V., Voronov, M.A., 2017. “Interpretation and use of measurements of the rangefinder sensor in the control tasks of the autonomous robot”. Upravlausie sistemy i masiny, 1, pp. 26–34. (In Russian).
  5. Lowe David G. Method and apparatus for identifying scale invariant features in an image and use of same for locating an object in an image, US Patent 6,711,293, March 23, 2004.
  6. Lowe David G. Object recognition from local scale-invariant features. International Conference on Computer Vision, Corfu, Greece (September 1999). 1999. pp. 1150–1157.
  7. Sukhoruchkina, O.M., Zakharov, A.O., 2010. “On object identifying technique using visual and sensory data of the intelligent mobile platform”. Pratsi Vseukrainskoi mizhnarodnoi konferentsii “UkrOBRAZ-2010”, Kyiv, pp. 11–14. (In Ukrainian).
  8. Iliin, S.O., 2007. “Special features of software implementation of the mobile robot control system”. Upravlausie sistemy i masiny, 4, pp. 28–42. (In Ukrainian).

Received 04.12.18