A device projects a virtual three-dimensional map of a physical environment onto an area captured by an image sensor of the device. The device receives user input to begin tracking device motion. In response to receiving the user input, the device tracks its motion within the area. The device determines location data within the area and orientation data of the image sensor within the area based on the tracked motion of the device. At least one of the location data or the orientation data is converted to real-world coordinates based on the virtual three-dimensional map of the physical environment. Path data is generated for a physical vehicle based on the at least one converted location or orientation data.