Marelli Europe S.p.a.

Davide Fontana

Axel Furlan

Erika Di Stefano

Enver Sangineto

Niculae Sebe

Ivan Chernukha

A system for obtaining a prediction an action (a) of a vehicle (V), including a camera for acquiring a sequence of images (F) a convolutional neural network visual encoder which obtains a corresponding visual features vector (v), one or more sensor that obtains a position of the vehicle (s) at the same time step (s), a Recurrent Neural Network configured to receive the visual features vector (v) and position of the vehicle (s) at the time step (t) and to generate a prediction of the action (a) of the vehicle (V). The system comprising a command conditioned switch configured upon reception of a control command (c) to select a corresponding branch of the Recurrent Neural Network. The system is configured to operate the selected corresponding branch to process the visual features vector (v) and position of the vehicle (s) at the time step (t) to obtain the prediction of the action (a) of the vehicle (V).

Growing community of inventors

Corbetta, Italy

Davide Fontana