Modern technologies make it possible today to compete autonomous cars (Self-driving cars) equipped with sensors and navigation, as well as using systems blockchain. XAIN - "eXpandable AI Network" is a research project University of Oxford i Imperial College in London, run by Leif-Nissen Lundbaka, who is today the CEO and prof. Michael Huth, current technical director.
XAIN AG, thanks to the use of modern solutions, ensures greater security of storage of acquired data, which is ordered in the projects of modern companies.
In 2017 year Xaine won Porsche Innovation Contest, which eventually led to cooperation with the manufacturer of these cars.
It aims to improve driving comfort and safety through use artificial intelligence.
The use of blockchain technology, originally developed for the financial industry, is an innovation here.
Assuming that unmanned vehicles must be able to adapt to the behavior of individual drivers, as well as have accurate navigation, operating in real time, it is necessary to prepare appropriate algorithms that process large amounts of data.
Modern autonomous vehicles must therefore monitor driver behavior, which is later analyzed.
So there were concerns about the security of collecting and storing such individual data.
Artificial intelligence companies reassure you that all collected data is anonymously collected, securely stored and saved in the model.
However, recent situations related to e.g. with the security of storing data (photos) of smartphones in the FaceApp application, used for facial recognition, show that the security features used so far are insufficient.
Xaine therefore, to protect privacy Federated Machine Learning (FedML), where data is stored in separate artificial intelligence models, aggregated and used later in the learning process.
The challenge is maximum simplification of the training process, which will ensure the security of the data obtained.
In addition to the fact that FedML simplifies the training process and ensures data security, it significantly reduces the cost of the entire enterprise.
Application ANDY supporting technology training models FedML XAIN is a system for automatic processing of item invoices.
For the first time, distributed systems were used to analyze the operation of controllers.
The company ensures that the system complies with data protection regulations (GDPR).
Not only Porsche, but also other manufacturers of self-propelled cars, including General Motors, BMW and Daimle also use distributed technologies in their research.
Porsche plans to use an artificial intelligence model to help drivers:
- adapting speed to current road conditions
- planning service and maintenance
- raising awareness of road hazards
- dynamically adjust the conditions inside the car, depending on who drives the car
- improving the visibility of all vehicles on the road
- driving planning (traffic jam assistant, automated parking)