Methods and systems for threat detection in vehicles
AssigneeZF Friedrichshafen AG
InventorsDeepashree Chicklore Mahesh, Max Himmel, Kalash Jain, Tammer Jarjoura, Matthew Jazwiecki, Mark Lundgren, Pulkit Monga, Srinivas Mysore, David Wilson, Grigory Yezersky
Life of this patent
Estimating the term…
Abstract
A method for identifying hazardous conditions during vehicle operation includes: accessing data from a sensor corresponding to images of one or more other vehicles; computing, with a machine-learned model, a threat estimate for the one or more other vehicles based at least in part on the data from the sensor; and transmitting a threat alert to a driver interface of an ego vehicle.
Claims18
A method for identifying hazardous conditions during vehicle operation, comprising:
- accessing, with a computing device on an ego vehicle, data from a sensor corresponding to images of one or more other vehicles;
- computing, with a machine-learned model on the computing device, a threat estimate for the one or more other vehicles based at least in part on the data from the sensor, the machine-learned model has been trained using a training dataset determined using information describing previous threats representing a plurality of threating vehicle behaviors, the training dataset comprising one or more positive samples, each positive sample representing a threating vehicle behavior of the plurality of threating vehicle behaviors, the machine-learned model being configured to identify from the data from the sensor one or more threating vehicle behaviors from the plurality of vehicle behaviors and to compute the threat estimate based on the one or more threating vehicle behaviors identified;
- and
- transmitting, with the computing device, data corresponding to a threat alert to a driver interface of the ego vehicle.
A system for identifying hazardous conditions during vehicle operation, comprising:
- an ego vehicle;
- a driver interface located on the ego vehicle;
- a sensor located on the ego vehicle;
- one or more processors located onboard the ego vehicle;
- and
- one or more non-transitory computer-readable media that store instructions that are executable by the one or more processors to perform operations, the operations comprising
- accessing data from the sensor corresponding to images of one or more other vehicles,
- computing, with a machine-learned model, a threat estimate for the one or more other vehicles based at least in part on the data from the sensor, the machine-learned model has been trained using a training dataset determined using information describing previous threats representing a plurality of threating vehicle behaviors, the training dataset comprising one or more positive samples each positive sample representing a threating vehicle behavior of the plurality of threating vehicle behaviors, the machine-learned model being configured to identify from the data from the sensor one or more threating vehicle behaviors from the plurality of vehicle behaviors and to compute the threat estimate based on the one or more threating vehicle behaviors identified, and
- transmitting data corresponding to a threat alert to the driver interface.
Dependencies are inferred from the claim text.