Autonomous driving is the future, but is it safe? With current technology, driving darkness or adverse weather conditions such as rain, fog and snow could be dangerous. The AUTOVISION Spearhead Project, announced by the Graphene Flagship, is developing a new high-resolution image sensor for autonomous vehicles, which can detect obstacles and road curvature even in extreme and difficult driving conditions.
Currently, self-driving cars use visible cameras, but in dense fog, these cameras are insufficient. Autonomous cars will also use LIDAR sensors, relying on pulsed laser to measure distances and constantly scan the area around them. However, this is a relatively slow-processing technology in comparison with the potential of new-generation imaging systems.
The AUTOVISION project is led by ICFO in Barcelona, and counts on the collaboration of industrial partners such Aixtron in the UK and Veoneer in Sweden, this new project will help make safe deployment of autonomous vehicles possible. Afterall, the success of autonomous driving will largely depend on how split-second moments of imminent hazards are handled.
The AUTOVISION project, over the course of three years, will produce complementary metal-oxide semiconductor (CMOS) graphene quantum dot image sensors in prototype sensor systems, ready for uptake in the automotive sector. Across the duration of the project, the developing image sensor is set to take huge leaps in sensitivity, operation speed and pixel size.
In the wider electronics industry, CMOS are at the heart of technological revolution. They have enabled compact and low cost micro-electronic circuits and imaging systems, but the diversification of this technology in applications other than microcircuits and visible light cameras has made limited progress. This is due to the difficulty of combining CMOS with other semiconductors apart from silicon.
Recently, monolithic integration of a CMOS integrated circuit with graphene has been made possible, enabling high-resolution image sensing that detects UV, visible, infrared and even terahertz frequencies.
The sensor's ability to see in the infrared — effectively night vision — means that same graphene CMOS sensors can be used as part of a self-driving car's automatic brake system, specifically in bad weather. This collision avoidance system is set to be a crucial application for graphene, and one that will help the wider uptake of autonomous driving technology.