Nasim Zoubeiri, Stephen Chui, Ryan Meghdies
supervised by Dr. Vadim Geurkov, ASSOCIATE PROFESSOR
Most people do not consider the sound of a car alarm to be a cause for concern. In fact, many of us use this feature to locate our vehicle in a parking lot. The issue with today's car alarms is not determining whether a vehicle has been broken into. Instead the issue lies in the before and after of the break-in scenario. That is what the focus of this project is and the problem this car security system hopes to solve. The system collects data from various sensors and uses an advanced algorithm to make intelligent decisions accordingly. We have integrated, what we call, "danger buffers" into our algorithm. These buffers are filled based on the data the system gathers. When these buffers pass a certain threshold the system will take action.
The system scans the perimeter of the vehicle looking for danger. If it feels threatened the system will either caution the threat using an audio message or contact the owner with pictures of the situation so that the human can make a decision on how to proceed. If the danger proceeds, the alarm will be set off. In the event that the alarm does go off, a text message will be sent to the vehicle owners' mobile device. Finally in the unfortunate even that the vehicle is stolen, the owner can track the vehicle on Google Maps via our web application. This web application may also be used by the owner, to remotely access data and control the system. All data is remotely loaded to an online server which allows the user to interact with the system from any device with internet connectivity.
Project targeted applications: Vehicle security, Remote system access and control, Advanced security systems