Traffic light planning is inherently difficult without accounting for pedestrians. Pedestrians congregating at an intersection can cause safety issues for all vehicles and people. Accurately determining the counts of pedestrians, vehicles, and bicycles is a difficult task for existing technologies in the market. The proposed technology allows for simple integration of counting of people, cars, and bikes and removes concerns that big brother is watching. Additional analysis can be performed by providing count data – if pedestrian counts are not changing, there may be a need to modify traffic patterns or alert first responders. Starting with simple, actionable data is the first step in this process, and will benefit the overall safety of the nation’s transportation system and ultimately strength the economy of the nation at large.

 

Our visual sensory module will observe the environment and, use its local processing unit analysis it. It will calculate the location of objects recognized and map it to the geographical information system. As the module consumes low-power, the solar panel is sufficient to power it. Also, the generated data is very rich and compressed, so it can be transferred using the NB-IoT network. There are two main innovations in this system. First, the low power computation unit can process footages locally without the need for a large amount of energy and a high-bandwidth network. Second, the mapping detected objects to geographical data with less preparation, which will reduce the installation cost significantly.