Back to results
US 11,772,670 B2Granted
Customized vehicle operation based on driver status
AssigneeMicron Technology, Inc.
InventorRobert Richard Noel Bielby
Life of this patent
Estimating the term…
Abstract
Embodiments of the present disclosure relate to assisted driving of a vehicle. The response time of the driver may be measured and control operations may be adjusted to accommodate or compensate the response time of the driver. Predictive actions in anticipation of a slightly delayed input from the driver may be taken to avoid potentially dangerous conditions. A vehicle may be controlled to slow down to allow more time to a potential event to accommodate a slow response from the driver.
Claims19
A system comprising:
- a sensor system configured to generate sensor data indicating proximity to objects;
- a memory that stores program instructions;
- and
- a processor configured to execute the program instructions to cause the processor to:
- analyze the sensor data to determine a braking condition based on a first proximity and velocity in relation to an object;
- set a driver status based on a machine learning model using a classifier that is trained according to training data, wherein the classifier classifies the braking condition, driver inputs, and timings of the driver inputs to determine whether a driver is alert;
- and
- update a braking system according to the driver status, wherein the braking system is configured to generate brake signals to cause a vehicle to decelerate, the braking system comprising a pedal configured to receive a driver brake input.
5 more elements
A system comprising:
- an interface to a braking system, the braking system being configured to generate brake signals to cause a vehicle to decelerate, the braking system comprising a pedal configured to generate at least some of the brake signals;
- a memory that stores program instructions;
- and
- a processor configured to execute the program instructions to cause the processor to:
- determine a level of driver alertness by analyzing at least one driver input in response to at least one condition, wherein the level of driver alertness is recorded as a driver status and wherein the condition is generated by a test system configured to measure driver reaction time;
- and
- adjust a configuration of the braking system according to the driver status, wherein determining the level of driver alertness comprises determining the level of driver alertness via a machine learning model using a classifier that is trained according to training data, wherein the classifier classifies the condition, the driver input, and timings of the driver input to determine whether a driver is alert.
4 more elements
A computer-implemented method comprising:
- sensing, by a sensor system, proximity to an object and a velocity in relation to the object;
- determining at least one condition based on the proximity and the velocity;
- detecting, by a processor, whether a driver provides a driver input within a predetermined window of time following the condition;
- setting, by the processor, a driver status based on a machine learning model using a classifier that is trained according to training data, wherein the classifier classifies the condition, the driver input, and timing of the driver input to determine whether the driver is alert;
- and
- activating an autonomous driving mode, by the processor, according to the driver status, wherein the autonomous driving mode is configured to generate a brake signal in response to sensor data indicating that at least the proximity to the object exceeds a threshold level.
3 more elements
Dependencies are inferred from the claim text.