Traffic Cones Detection for Autonomous Ground Vehicle

Rodi Hartono, Taufiq Nuzwir Nizar, Didit Andri Jatmiko, Feki Pangestu Wijaya
dx.doi.org/10.17576/apjitm-2022-1101-06

Abstract

Jelajah V-18 is an autonomous vehicle prototype where can navigate autonomously using Global Positioning System (GPS) for long-distance navigation and can find out the surrounding conditions to navigate short distances using cameras and proximity sensors. The vehicle also has a system that can search and recognize a static object using an image processing system. This research focuses on the detection of an objects in the form of traffic cones using the (Hue, Saturation, Value) HSV method to recognize objects by the color, then with the binary method the image of the object will be converted back to get the edges of the object shape and contour where the system can recognize the shape of the object. Based on the implementation and experiments conducted, the robot can detect traffic cone objects with a success rate of 100 percent at a distance of 1-6 meters and 35.67 percent at a distance of 7 meters.

keyword

Autonomous vehicle, Object Detection, HSV Method, Binary Method, Contour Detection

Area

Pattern Recognition