In self-driving platooning technology, which allows many vehicles to drive in clusters, it is very important to identify the path of the leader vehicle''s steering control. when optical camera communication technology is applied to platooning driving, the movement route of the leading vehicle can be effectively grasped by using the taillight of the leading vehicle and the camera of the rear vehicle. However, when the ROI area in the image of the taillight of the leading vehicle is not recognized or the taillight of the front vehicle is broken, the steering control can not be smooth. In this paper, the forward vehicle transmits the speed data, the type of vehicle, the taillight left side, and the right side indicator data to complement these problems and actively utilize it in the rear vehicle and lane recognition system increased the distribution of recognizable lanes, enabling lane recognition. In addition, in the situation where lane recognition is inevitable, we implemented a technology to control steering by considering the center point of lane using only the lane recognized. The results of the experiment showed that the platooning driving steering control was performed stably even in the situation where the ROI area of the left and right taillights was not recognized and the lane recognition was not realized. But when all algorithms are used in real vehicles to conduct experiments on real roads, Although the data is accurately received, the reception rate of the received data is significantly lower than the number of data transmitted. To solve this problem, the ROI area cannot be accurately detected each time, since the ROI area is detected based on differences between frames using YOLov2, one of the deep learning techniques used primarily in the field of image processing, and the ROI station for the entire frame, a drawback of the previous image processing technology, cannot be recognized. The proposed method implemented a more accurate and secure optical camera communication-based autonomous navigation platform communication system based on an improved reception rate than traditional imaging processing techniques, since the ROI area can be accurately detected by receiving the taillight''s left and right boundary box coordinates.