Technology is evolving fast and the automotive area is trying to evolve at a similar rhythm. The vehicles, in special the cars, are getting more complex and smarter but the interfaces between the driver and vehicle, the automotive HMI's (Human Machine Interfaces), tend to be less intuitive and easy to use with the increasing quantity and variety of the technological solutions in the vehicles.

One major problem, directly connected with non-intuitive and hard to use interfaces, is the driver's distraction. The driver's distraction is one of the major causes of traffic accidents. Globally, each year, around 1.25 million people die in traffic accidents, being driver's distraction the cause of almost 20% of all traffic accidents.

Many technologies and concepts are now emerging to solve or minimize the traditional HMI's problems, being gesture recognition one of those technologies and concepts.

Gesture recognition, applied to automotive interfaces, is a very interesting concept, mainly because it tries to join the functionalities of a "traditional" HMI with an easy to use interface, consisting mainly in a simple set of gestures instead of a huge amount of actions involving the "regular" buttons, dials, switches and screens.

This thesis is integrated in the project "INNOVCAR: The cockpit of the future" and have as purpose the study and implementation of alternative concepts of automotive HMI, that are contactless and eyes-free, using gesture recognition concepts. To achieve this objectives the gesture recognition system is based on infrared technology, mainly because of the flexibility and relation cost vs. benefit of this technology.

This thesis and project have as main contributions the development of a new base of knowledge, practical and theoretical, to be mainly shared with Bosch. This base of knowledge will consist in a functional gesture recognition system and a set of automotive HMI concepts using infrared technology. This base of knowledge is important, mainly because could be used to select new concepts and technologies for further investigation or could be used in the development of future products.

As conclusion, the thesis hopes to offer a clear vision about the possible uses of gesture recognition applied to automotive interfaces and also its benefits in terms of driver's performance, comfort and safety. The question of safety is very relevant and is expected that by reducing the driver's distraction, the number of crashes should fall as well as the possible number of dead's.

Finally, the improvement of the gesture detection algorithms, the development of new functionalities and concepts of HMI are possible improvements of this thesis.