Autonomy
In order to meet its purpose of autonomous parcel delivery, the EVs must be able to autonomously navigate the upper campus footpath network, including wheelchair ramps. In doing so, it will be necessary to detect and stop before coming into contact with potential obstacles such as pedestrians, cars, walls, etc. A static control center will be established for the purposes of mission control and COSMOS operation, which oversees the AEVS. The CC will be composed of a computer capable of running COSMOS and connected to the network. For monitoring and safety purposes, a HexCamera will be fitted to the top of the vehicle to provide a 360-degree live video stream to the CC.
LEVEL 3
In this iteration, the system's autonomy will be improved from Level 2 to Level 3.
At this level, the vehicle shall be able to navigate without human intervention except in case of an emergency. This entails detecting and stopping for obstacles, independent decision making, GPS location tracking, and a live video feed and connection for monitoring and assumption of control. This will be done from a central location responsible for overseeing the activity of all vehicles in the system. Sensors and program reaction time will need to be sufficient to stop the vehicle from a top speed of 4 mph upon obstacle detection. However, not all obstacles will be stationary. It will also be necessary to detect and stop for moving obstacles for which the ultrasonic sensor’s limited range does not provide adequate forewarning. To this end, You Only Look Once real-time object detection software may be used with the HexCamera to measure obstacle speeds. Those in the direction of travel exceeding a net speed (V_AEVS + V_observer) should require more stopping time. Obstacles moving in the same direction as the AEV will have a negative parallel velocity component, and vice-versa.