Driver behavior cloud-side collaborative learning system based on federated transfer learning

Abstract

The invention relates to a driver behavior cloud-side collaborative learning system based on federated transfer learning. The system comprises an AI cloud platform, an intelligent vehicle-mounted terminal, a deep neural network system for federated transfer learning and the like. The deep neural network system comprises a vehicle-mounted end neural network and a platform end neural network, and can realize federated model training and learning migration of non-original data transmission between the vehicle-mounted end neural network and the platform end neural network based on an AI cloud platform and intelligent vehicle-mounted terminals of a plurality of vehicles connected with the AI cloud platform. The method is used for intelligent automobile driver behavior perception and intention understanding, and compared with the prior art, the driver collaborative perception machine learning problem of a large-scale vehicle cluster is solved, and meanwhile driver privacy is protected.

Description

technical field [0001] The invention relates to the field of behavior perception and intention understanding of smart car drivers, in particular to a driver behavior cloud-edge collaborative learning system based on federated transfer learning. Background technique [0002] With the rapid integration of automobiles and artificial intelligence, 5G communications, the Internet, sensor technology and other fields, intelligent networked vehicles have become the main trend of future automotive technology development and will profoundly change the way people travel. Domestic and foreign auto groups, well-known auto technology research and development giants, and external companies in the fields of the Internet, communications, and electronics have deployed the intelligent networked auto industry to accelerate the transformation and implementation of commercial demand. Commercial scenarios such as park logistics, self-driving buses, fixed-line expressway freight, and online driverl...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to provide a driver behavior cloud-side cooperative learning system based on federated transfer learning to overcome the defects of the above-mentioned prior art, which solves the scientific problem of large-scale driver cooperative perception machine learning, and at the same time avoids infringement of personal The application problem of privacy is suitable for the application and promotion of driver behavior monitoring in large-scale online car-hailing and logistics fleet operations

Images