CASPIAN JOURNAL

MANAGEMENT AND HIGH TECHNOLOGIES

VIRTUAL DRIVER'S ASSISTANT FOR SEARCH OF A PLACE IN OPEN PARKING

Read

Kravets Alla G., Astankov Alexey A., Mokrushin Georgy A. VIRTUAL DRIVER'S ASSISTANT FOR SEARCH OF A PLACE IN OPEN PARKING // Caspian journal : management and high technologies. — 2022. — №2. — pp. 35-44.

Kravets Alla G. - Volgograd State Technical University, Volgograd, Russian Federation

Astankov Alexey A. - Volgograd State Technical University, Volgograd, Russian Federation

Mokrushin Georgy A. - State University "Dubna", Dubna, Russian Federation

The article discusses the development of a virtual assistant for searching a place in an open parking lot. Most often, when arriving at shopping centers and shops, the driver cannot find a place in open parking lots, and he has to look for it for a long time. This entails an increase in environmental pollution and an increase in the number of accidents. Or drivers have to park their car in the wrong place (carriageway, sidewalk), which leads to incomplete use of the roadway and discomfort to passers-by. The proposed approach helps to solve this problem by quickly navigating to a free parking space in nearby outdoor parking lots. During the study, modern approaches, analogues of systems, were analyzed. A virtual assistant was implemented to simplify the parking process. To test the assistant, images from streaming video cameras were used. The assistant design process, its functions and capabilities were described. The assistant server performs the functions of storing parking markings, parking space status, user information, and free parking spaces are detected on the server using images from streaming video cameras and the MASK R-CNN neural network. The assistant client performs the functions of displaying free parking spaces and navigating to them.

Key words: open parking, MASK R-CNN, virtual assistant, neural network, Yandex.Maps