Method for Building an Environment Model of Autonomous Vehicles

J.A. Lyakhovenko, S. Mukhamedzhanov, I.I. Komarov

Abstract


An environment model describes a set of abstractions designed to represent all significant physical and virtual elements in space. The use of asset models is especially important when autonomous vehicles (AVs) encounter unforeseen circumstances or technical failures. In the proposed method for constructing a model of environment, the AV first collects all information about objects in the external environment, after which it conducts cluster analysis to reduce them to certain types. Then it determines possible states and behavior scenarios for each of the defined types. In the process of further research, the vehicle uses technical vision systems to detect an object and, based on certain characteristics, classify it as the closest type with a certain threshold (distance from the cluster). If such an operation is impossible, a new type is created.

The presented method for constructing a space model determines the automated collection and description of information about the external environment, which makes it possible to increase the efficiency of autonomous vehicles due to the possibility of continuous updating of the space model. In the future, it is planned to refine the model in terms of determining the degree of confidence in the new information received.

 


Full Text:

PDF (Russian)

References


Article: Self-driving cars (world market). Gartner - statistics on BTS. [Electronic resource]. – URL: https://www.tadviser.ru/index.php/

Road safety indicators. State traffic inspectorate. [Electronic resource].– URL: http://stat.gibdd.ru/

Min H. et al. Kinematic and dynamic vehicle model-assisted global positioning method for autonomous vehicles with low-cost GPS/camera/in-vehicle sensors //Sensors. – 2019. – Т. 19. – №. 24. – С. 5430.

Li S. et al. Planning and decision-making for connected autonomous vehicles at road intersections: A review //Chinese Journal of Mechanical Engineering. – 2021. – Т. 34. – С. 1-18.

Li Z. et al. Obstacle Avoidance Trajectory Planning for Autonomous Vehicles on Urban Roads Based on Gaussian Pseudo-Spectral Method //World Electric Vehicle Journal. – 2023. – Т. 15. – №. 1. – С. 7.

Feng A. et al. Efficient Autonomous Exploration and Mapping in Unknown Environments //Sensors. – 2023. – Т. 23. – №. 10. – С. 4766.

Yamauchi B. A frontier-based approach for autonomous exploration //Proceedings 1997 IEEE International Symposium on Computational Intelligence in Robotics and Automation CIRA'97.'Towards New Computational Principles for Robotics and Automation'. – IEEE, 1997. – С. 146-151.

Wu C. Y., Lin H. Y. Autonomous mobile robot exploration in unknown indoor environments based on rapidly-exploring random tree //2019 IEEE International Conference on Industrial Technology (ICIT). – IEEE, 2019. – С. 1345-1350.

Dang T. et al. Explore locally, plan globally: A path planning framework for autonomous robotic exploration in subterranean environments //2019 19th International Conference on Advanced Robotics (ICAR). – IEEE, 2019. – С. 9-16.

Selin M. et al. Efficient autonomous exploration planning of large-scale 3-d environments //IEEE Robotics and Automation Letters. – 2019. – Т. 4. – №. 2. – С. 1699-1706.


Refbacks

  • There are currently no refbacks.


Abava  Кибербезопасность MoNeTec 2024

ISSN: 2307-8162