Ford Global Technolgoies, LLC

Kun Deng

Alex Maurice Miller

Nanjun Liu

Wangdong Luo

Roger Arnold Trombley

Ahsan Qamar

Thomas Edward Pilutti

Gerald H Engelman

Sarra Awad Yako

Timothy Drotar

Fangjun Jiang

Jerry H Engelman

Gary Song

Patrick Piastowski

A system includes a computer including a processor and a memory, the memory storing instructions executable by the processor to determine respective threat numbers for each of a plurality of targets based on an angular acceleration of a host vehicle and actuate a component in the host vehicle based on the threat numbers.

Collision avoidance and mitigation

A distance offset is determined based on a determined time to collision, a relative lateral distance, and a relative longitudinal distance between the target and a host vehicle. A threat estimation is determined based on the distance offset and a distance threshold. A component of the host vehicle are actuated based on the threat estimation.

Collison mitigation and avoidance

A respective confidence level of a potential collision is determined for each of a plurality of targets based on each target's heading angle and distance from a host vehicle. A threat number is determined for each target when its respective confidence level is above a threshold. A vehicle component is actuated based on the threat number.

Collision mitigation and avoidance

A plurality of targets are identified. A path for each target is predicted. A threat number for each target is determined based at least in part on the predicted paths. The threat number indicates a probability of a collision between the respective target and a host vehicle. One or more vehicle subsystems in the host vehicle is actuated based on the threat numbers.

Vehicle collision avoidance

A time to collision is determined between a turning host vehicle and each of a plurality of targets. Based on a lateral distance and a longitudinal distance, a minimum distance is determined between the turning host vehicle and each of the plurality of targets. A threat number is determined for each target selected based on the time to collision and the minimum distance. A vehicle component is actuated based on the threat number.

A controller in a host vehicle includes a processor and a memory storing processor-executable instructions. The processor is programmed to calculate an error bound for confidence intervals of a predicted position at a future time of a target vehicle relative laterally to a current position and orientation of the host vehicle at a current time based on a current position and velocity of the target vehicle and the future time.

Autonomous vehicle using path prediction

A controller includes a processor and a memory storing processor-executable instructions in a host vehicle. The processor is programmed to designate a target vehicle as a collision risk upon determining that an intersection angle between the host vehicle and the target vehicle is between preset lower and upper bounds and an absolute value of an azimuth angle rate of change of the host vehicle is below a preset threshold.

Collision-warning system

An intersection of a host vehicle and a target vehicle is identified. Data relating to the target vehicle are collected. A map of a surrounding environment is developed. A driver intent probability is determined based at least in part on the map. A threat estimation is determined based at least in part on the driver intent probability. At least one of a plurality of safety systems is activated based at least in part on the threat estimation.

Growing community of inventors

Ann Arbor, MI, United States of America

Kun Deng