US 11,055,605 B2
Detecting dangerous driving situations by parsing a scene graph of radar detections
Hans Peter Graf, Lincroft, NJ (US); Eric Cosatto, Red Bank, NJ (US); and Iain Melvin, Hopewell, NJ (US)
Assigned to NEC Corporation
Filed by NEC Laboratories America, Inc., Princeton, NJ (US)
Filed on Oct. 17, 2017, as Appl. No. 15/785,796.
Claims priority of provisional application 62/489,539, filed on Apr. 25, 2017.
Prior Publication US 2018/0307967 A1, Oct. 25, 2018
Int. Cl. G06N 3/08 (2006.01); G06N 3/04 (2006.01); G01S 7/41 (2006.01); G01S 13/931 (2020.01)
CPC G06N 3/04 (2013.01) [G01S 7/417 (2013.01); G06N 3/08 (2013.01); G01S 13/931 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A computer-implemented method executed on a processor for training a neural network to recognize driving scenes from sensor data received from vehicle radar, the method comprising:
extracting substructures from the sensor data received from the vehicle radar to define a graph having a plurality of nodes and a plurality of edges, each of the substructures being a path from an initial node through one or more detected nodes up to a maximum predetermined depth;
constructing, by a construction module, a neural network for each extracted path, an initial point of each constructed neural network being a NULL bias vector learned during training;
concatenating the outputs of each of the constructed neural networks for each of the plurality of edges into a single vector describing a driving scene of a vehicle; and
classifying, by a classifying module, the single vector into a set of one or more driving scene situations involving the vehicle represented as a vector of numbers, each number corresponding to a probability of a particular driving scene situation being present in the driving scene.