SaferAuto is a modular driver assistance system. Between its features, it would detect road elements such as traffic lights, traffic signs and pedestrians. Besides, it would inform the driver of dangerous environments or situations, reducing the risk of having an accident. Finally, it would have an integrated chatbot with the ability to interact with the audio in the car, answer questions regarding the weather or informing of the best path to reach the destination.
The modularity feature involves developing the project in a way that future skills can be added to the system in an easy way, adapting the system to the needs of a customer or company. An example of these future capabilities could be the detection of animals, the road lane and so on.
It is important to highlight several features the system has to possess in order to be a real and useful product. Firstly, it must not only detect the road elements but do so in real-time, in order that the information can be processed faster than a human. Secondly, apart from detecting the elements, it would have to track them and filter which ones are important to the road lane the car is in. Finally, the accuracy of the detection must be high, in order to avoid accidents instead of causing them.
Regarding the installation process, it would be really easy. The software would be installed in an embedded device with a camera, situated under the rear-view mirror.
Developing systems of autonomous driving is a hot topic nowadays. Companies such as Google, Uber and Tesla are investing a lot of money in those systems, in order to pursue an ideal future of driverless cars. In respect to the motivation of the project, it would help during the transition from normal cars to autonomous ones, reducing the risk of having accidents while driving by detecting dangerous situations much faster than a human. That is why I believe it is important to make this project open-source, in order to invite other developers to contribute to this project thanks to its modularity feature.
Ángel Igareta - Computer Engineering Student