Processing

Please wait...

Settings

Settings

Goto Application

1. WO2020091835 - GENERATING TESTING INSTANCES FOR AUTONOMOUS VEHICLES

Publication Number WO/2020/091835
Publication Date 07.05.2020
International Application No. PCT/US2019/024952
International Filing Date 29.03.2019
IPC
G06N 20/00 2019.01
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
20Machine learning
G06N 3/08 2006.01
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
3Computer systems based on biological models
02using neural network models
08Learning methods
CPC
G05B 13/0265
GPHYSICS
05CONTROLLING; REGULATING
BCONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
13Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
02electric
0265the criterion being a learning criterion
G05D 1/0088
GPHYSICS
05CONTROLLING; REGULATING
DSYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
1Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
0088characterized by the autonomous decision making process, e.g. artificial intelligence, predefined behaviours
G05D 1/0246
GPHYSICS
05CONTROLLING; REGULATING
DSYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
1Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
02Control of position or course in two dimensions
021specially adapted to land vehicles
0231using optical position detecting means
0246using a video camera in combination with image processing means
G05D 1/0257
GPHYSICS
05CONTROLLING; REGULATING
DSYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
1Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
02Control of position or course in two dimensions
021specially adapted to land vehicles
0257using a radar
G05D 1/027
GPHYSICS
05CONTROLLING; REGULATING
DSYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
1Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
02Control of position or course in two dimensions
021specially adapted to land vehicles
0268using internal positioning means
027comprising intertial navigation means, e.g. azimuth detector
G05D 1/0278
GPHYSICS
05CONTROLLING; REGULATING
DSYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
1Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
02Control of position or course in two dimensions
021specially adapted to land vehicles
0276using signals provided by a source external to the vehicle
0278using satellite positioning signals, e.g. GPS
Applicants
  • AURORA INNOVATION, INC. [US]/[US]
Inventors
  • VALOIS, Jean-Sebastien
  • MUNOZ, Daniel
Agents
  • SALAZAR, John, F.
  • EICHENBERGER, Robert, H.
  • STINEBRUNER, Scott, A.
Priority Data
16/271,40108.02.2019US
62/755,19702.11.2018US
Publication Language English (EN)
Filing Language English (EN)
Designated States
Title
(EN) GENERATING TESTING INSTANCES FOR AUTONOMOUS VEHICLES
(FR) GÉNÉRATION D'INSTANCES DE TESTS POUR VÉHICULES AUTONOMES
Abstract
(EN)
Sensor data collected from an autonomous vehicle (100, 502, 602, 702, 802) can be labelled using sensor data collected from an additional vehicle (282, 506, 606, 706,806). Labelled sensor data can generate targeted testing instances for a trained machine learning model, where the trained machine learning model is used in generating control signals for an autonomous vehicle (100, 502, 602, 702, 802). In many implementations, targeted training instances (316) can generate an accuracy value for the trained neural network model (312, 404). Additionally or alternatively, the sensor suite on the additional vehicle (282, 506, 606, 706, 806) can include a removable hardware pod (202, 508, 608, 708, 804) which can be mounted on a variety of vehicles.
(FR)
Selon la présente invention, des données de capteur collectées à partir d'un véhicule autonome (100, 502, 602, 702, 802) peuvent être marquées à l'aide de données de capteur collectées à partir d'un véhicule supplémentaire (282, 506, 606, 706, 806). Des données de capteur marquées peuvent générer des instances de test ciblées pour un modèle d'apprentissage machine entraîné, le modèle d'apprentissage machine entraîné étant utilisé pour générer des signaux de commande pour un véhicule autonome (100, 502, 602, 802). Dans de nombreux modes de réalisation, des instances d'apprentissage ciblées (316) peuvent générer une valeur de précision pour le modèle de réseau neuronal entraîné (312, 404). De plus ou en variante, la suite de capteurs sur le véhicule supplémentaire (282, 506, 606, 706, 806) peut comprendre un module matériel amovible (202, 508, 608, 708, 804) qui peut être monté sur une variété de véhicules.
Latest bibliographic data on file with the International Bureau