Errors in Measuring the Distance to an Obstacle by Technical Vision Means and in Forecasting Braking Distance in Driverless Train Control Systems. World of Transport and Transportation

Technical vision systems are sources of information about an obstacle on the track in the case of driverless train control. Based on the information received, the traffic control system decides to turn on the braking mode to prevent a colliosni with an obstacle. In accordance with international and domestic expertise and standard ratings, it is necessary to ensure the probability of a dangerous failure, in this case, the probability of hitting an obstacle, not more than 10-8 with a confidence probability of 0,95 according to SIL-4 ([Russian state standard] GOST-R61508). Considering the presence of an error in measuring the distance to an obstacle by the technical vision system and an error in calculating the stopping distance, it is required to determine the coordinate of the braking start point when an object is detected on the track in such a way as to ensure that the train stops before the obstacle with a probability determined in accordance with SIL-4.

A feature of the problem being solved for estimating the errors in measuring the distance to an obstacle and calculating the stopping distance implies the need to determine the estimates of their maximum values and to develop an algorithm for using these estimates in such a way that the collision probability does not exceed the normalised value.

A technique is described for determining the maximum value of the error in measuring the distance to the obstacle, the probability of exceeding which is quite small (from 10-2 to 10-6). A proposed algorithm for multiple measurements of the distance to an obstacle allows choosing the minimum measurement result for deciding on the start of braking, which ensures meeting standard indicator of a probability of a train colliding with an obstacle according to SIL-4. A method for estimating the error in calculating the stopping distance has been developed, which, together with the algorithm of multiple measurements by the technical vision system of the distance to the obstacle, provides the standard indicator according to SIL-4. The need for the second channel of technical vision due to the presence of curves along the route is shown. The necessity of using algorithms for multiple measurements to an obstacle through the second channel located outside the train is also substantiated. It is noted that the methods described in this article for choosing the maximum values of random errors in measurements and calculations, the values of which can be exceeded with a very low probability, can be used to solve various applied problems of traffic control in transportation processes.

1. Matthies, Larry Obstacle Detection, 2014 In: Ikeuchi, K (eds) Compouter Vision Springer, Boston, Ma DOI: 10 1007/978-0-387-31439-6_52

2. Jain, R , Tamgade, P , Swaroopa, R , Bhure, P , Shahu, S , Pote, R Simulation of Obstacle Detection of an Autonomous Car International Journal of Advanced Research in Science, Communication and Technology, 2021, pp 430−435 DOI: 10 48175/IJARSCT-1420

3. Asuka, Masashi; Kataoka, Kenji; Komaya, Kiyotoshi; Nishida, Syogo Automatic Train Operation Using Autonomic Prediction of Train Runs IEEJ Transactions on Industry Applications, 2008, Vol 128, pp 1365–1372 DOI: 10 1541/ieejias 128 1365

4. Chen Zhang; Xuewu Xu; Chen Fan; Guoping Wang Literature Review of Machine Vision in Application Field E3S Web of Conferences, 2021, Vol 236, pp 04027 DOI: 202123604027

5. Zhongfei Zhang, Weiss, R , Hanson, A R Obstacle detection based on qualitative and quantitative 3D reconstruction. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1997, Vol 19, pp 15−26 DOI: 10 1109/34 566807

6. Feiden, D , Tetzlaff, R Cellular neural networks for motion estimation and obstacle detection Advances in Radio Science, 2003, Vol 1, pp 143–147 DOI: 10 5194/ars-1-143-2003

7. Lourenço, A , Marques, F , Santana, P , Barata, J A Volumetric Representation for Obstacle Detection in Vegetated Terrain, 2014 IEEE International Conference on Robotics and Biomimetics, IEEE ROBIO 2014 DOI: 10 1109/ROBIO 2014 7090344

8. Bernini, N , Bertozzi, M , Castangia, L , Patander, M , Sabbatelli, M Real-Time Obstacle Detection Using Stereo Vision for Autonomous Ground Vehicles: A Survey, 2014 17th IEEE International Conference on Intelligent Transportation Systems, ITSC 2014 DOI: 10 1109/ITSC 2014 6957799

9. Takahashi, Katsuhiko Obstacle detection device and method and obstacle detection system, 2014 [Electronic resource]: https://www researchgate.net/publication/302747500_Obstacle_detection_device_and_method_and_obstacle_detection_system Lastaccessed 16 11 2021

10. Khan, Umair; Fasih, Alireza; Kyamakya, Kyandoghere; Chedjou, J Genetic Algorithm Based Template Optimization for a Vision System: Obstacle Detection, 2009 [Electronic resource]: https://www.researchgate net/publication/228347428_Genetic_Algorithm_Based_Template_Optimization_for_a_Vision_System_Obstacle_Detection/ Last accessed 16 11 2021

11. Okhotnikov, A L , Chernin, M A Development of systems for autonomous rolling stock [Razrabotka sistem dlya avtonomnogo podvizhnogo sostava] Avtomatika, svyaz, informatika, 2001, Iss 11, pp 21−24 DOI: 10 34649/AT 2021 11 11 006

12. Ventzel, E S Probability Theory: Textbook [Teoriya veroyatnostei: Uchebnik] 12th ed , ster Moscow, Yustitsiya publ , 2018, 658 p ISBN 978-5-4365-1927-2

13. Baranov, L A Evaluation of Metro Train Succession Time for Safety Systems Based on Radio Channel World of Transport and Transportation, 2015, Vol 13, Iss 2, pp 6−24 [Electronic resource]: https://mirtrelpub ru/jour/article/view/260 Last accessed 16 11 2021

14. Baranov, L A , Golovicher, Ya M , Erofeev, E V , Maksimov, V M Microprocessor-based automatic control systems for electric rolling stock [Mikroprotsessornie sistemy avtovedeniya elektropodvizhnogo sostava] Ed by Baranov, L A Moscow, Transport publ , 1990, 272 p ISBN 5-277-00964-7

15. Bestemyanov, P F Methods for improving safety of microprocessor systems for interval regulation of train traffic Abstract of D Sc (Eng) thesis [Metody povysheniya bezopasnosti mikroprotsessornykh sistem intervalnogo regulirovaniya dvizheniya poezdov. Avtoref. dis… dok. tekh. nauk] Moscow, MIIT publ , 2001, 48 p

16. Nikiforov, B D , Golovin, V I , Kutiev, Yu G Automation of train traffic control [Avomatizatsiya upravleniya dvizheniem poezdov] Moscow, Transport publ , 1985, 263 p

17. Pudovikov, O E , Kiselev, M D Optimization of Parameters of Automatic Speed Control System of a Freight Train with Distributed Traction Russian Electrical Engineering, 2020, Vol 91, No 9, pp 568−576 [Electronic resource]: https://elibrary ru/item asp?id=45136886 Last accessed 16 11 2021

18. Batenko, A P Control of the finite state of moving objects [Upravlenie konechnym sostoyaniem dvizhushchikhsya ob’ektov] Moscow, Sov Radio publ , 1977, 256 p.