Open AccessDesign and implementation of obstacles detection in self-driving car prototype
Author(s) -
Rizky Sari,
Ig. Prasetya Dwi Wibawa,
Cahyantari Ekaputri
Publication year - 2019
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1367/1/012020
Subject(s) - obstacle , automotive industry , computer science , artificial intelligence , computer vision , self driving , path (computing) , automotive engineering , engineering , geography , archaeology , aerospace engineering , programming language
Technology progresses very rapidly that makes a variety of industries vying to produce new innovations, especially the automotive industry. One example of innovation in the automotive industry is a self-driving car that can drive itself. Therefore, a self-driving car prototype is needed. This research focuses on the obstacles detection and calculation of obstacle distance. The obstacle detection uses a webcam with digital image processing. Obstacles which are detected are quiescent blue object that blocks the path of a self-driving car prototype and red light of traffic lights. The results of this research have a success rate at distance from webcam to object with actual distances of 38,3 cm and 55 cm amounting to 85,9% and 92,83%, the success rate at distance from webcam to blue obstacle with actual distances of 37,8 cm and 44,5 cm are 91,865% and 97,744%, the test results of red HSV value have accuracy 80%, precision 76,47%, and recall amounting to 86,667%, and the test results of blue HSV value have accuracy 83,33%, precision 81,25%, and recall 86,667%.