The Safe Autonomous Land Transport (TAS) project will aim to automate functions such as distinguishing between lineside signs and signals, spotting passengers on a station platform, monitoring defects on passing trains, and recognising objects on the track. The objective is to carry out these functions to very high levels of safety. TAS will also work with Compiègne Technical University (UTC) to draw on its expertise.
An environmental perception system will be designed based on a combination of sensors, such as radar and Lidar cameras, to provide advanced functionalities based on artificial intelligence.
TAS will provide proof of concept through laboratory modelling which will test the choice of sensors used and carry out virtual validation using an iterative approach. This will be followed by tests on an SNCF locomotive which will be conducted between Villeneuve Saint-Georges and Monterau.
“At SystemX, our teams have recognised expertise in the field of artificial intelligence and methodologies for the processing of images applied to transport, as well as expertise in virtualisation and simulation to validate and certify the security of these types of systems,” says Mr Eric Perrin-Pelletier, managing director of SystemX. “Our agility and speed of execution will enable us to produce an automaton in less than 18 months, then carry out its functional demonstration on site within 24 months at most.”
Mr Pierre Izard, SNCF’s chief technology and rail systems officer, says the project will enable SNCF to prepare for the “train of the future.”
IRT SystemX was launched in February 2012 as part of the French government’s Investment for the Future programme to support innovation in France.