Nowadays, Global Positioning Systems (GPS) are used everywhere for positioning and navigation. However, its use is not suitable in indoor environment, due to power budget constraints and the strong attenuation inside buildings. Therefore, indoors navigation takes advantage of other technologies to infer position. Recently, several Visible Light Positioning (VLP) systems have been reported. Among these technologies, Visible Light Communication (VLC) is one of the most promising, as its operation is based on the use of LED lights, currently widely used in the illumination solutions of most buildings. In this paper, we propose an indoor navigation system based on VLC in an industrial application for automated warehouses, where the navigation of autonomous vehicles (AVG) is supported by VLC. The proposed VLC system establishes bidirectional communication between the infrastructure and the guided vehicles. LED transmitters at the warehouse ceiling support downlink data transmission from the Infrastructure to Vehicle (I2V). This channel provides positioning and navigation of the vehicles, as well as transmission of dedicated messages related to the requested tasks of the management warehouse system to the autonomous vehicles. The uplink channel from the Vehicle to the Infrastructure (V2I) is used to acknowledge the requested tasks and transmit updates on the concluded tasks. Optical transmitters are tri-chromatic white LEDs with a wide angle beam. The characterization of the optical transmitter system is done through MatLab simulations for path loss and VLC channel gain prediction, using the Lambertian model for the LED light distribution. Dedicated receivers based on a-SiC:H/a-Si:H photodiodes with selective spectral sensitivity are used to record the transmitted signal. The decoding strategy is based on accurate calibration of the output signal.