Result description
Solution based on computer vision and deep learning for the automated drowsiness detection in drivers of vehicles, for a satefy driving. Two types of solutions are proposed:
- A full embedded solution which involves a set of hardware components to be fitted inside of the vehicle with a software installed in the processing unit.
- A embedded software solution to be integrated in a compatible equipment of the client, fulfiling a list of specifications.
Our DMS, based on AI computer vision, is designed to monitor the driver to avoid accidents that, in many cases, can be fatal. The system processes the images locally and not in the Cloud. For this we have developed different layers of artificial intelligence algorithms based on convolutional neural networks that detect fatigue and drowsiness (closing eyes, yawning, head tilt, etc.) in drivers, warning them before it is too late.
The system has also been tested to work with autonomous vehicles and to help decision-making at those levels where the vehicle needs driver interaction.
Addressing target audiences and expressing needs
- Expanding to more markets /finding new customers
- Business Angels
- Venture Capital
Nowadays, the business plan of the solution has two options, according to the roadmap defined:
- Evolve the solution, adding new functionalities and new integrations related to the current existing hardware which are working OEMs and TIER-1s/TIER-2s companies. In this option, fiding potencial clients or stakeholders will be an important part of the actions to do.
- Find potencial investors, etc. to add new types of investments to speed up the launch to market.
- Public or private funding institutions
- Other Actors who can help us fulfil our market potential
- Private Investors
R&D, Technology and Innovation aspects
The current stage of the solution is in the validation process in a relevant environment. The DMS will be part of decision-making process in some use cases, being feedback, according to the levels of Driving Automation for On-Road Vehicles.
Regarding the next steps, described previously, the evolution of the prototype will be carried out and the search of potencial clients and investors. On the other hand, new driver’s behaviour patterns are going to be developed as new layers of artificial intelligence algorithms, in order to detect the identity of the driver (to prevent theft, customize the vehicle settings to each driver, unfit or unauthorized driver), smoking drivers or the use of mobile phones (useful for companies transporting dangerous goods, logistic companies, public transport, rental cars, insurance companies, etc).
The solution is moduled to an easy integration, keeping the same software for a full solution, or with minor modifications in the case of adding new hardware to the system.
From the point of view of the software, n the case of the business case related to the software integration in others external platforms, the replicability will require some adaptions in the first stages of the process, but totally automated in the future.
Nowadays, the driver monitoring system is being part of the novel safety components integrated in the vehicles, and it will a mandatory safety system in the future cars, such as: the safety belt, the airbag or the ABS breaks.
The main objective ow is to establish joint agreements with car manufactures (OEMs) and technology providers (TIER1 and TIER2) to be part of the evolution cycle of this technology.
Result submitted to Horizon Results Platform by ROVIMATICA SL
