IP Library Granted Patent US 9,440,650
Granted Patent B2
US 9,440,650 · App. 14/409,980 · Granted Sep 13, 2016

Collision prediction apparatus

Inventor: Takeshi Nanami (Gotenba, JP)
Assignee: TOYOTA JIDOSHA KABUSHIKI KAISHA
B60W30/095B60R21/01B60W40/04B60W2550/302B60W2550/306
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 9,440,650
App. No.
14/409,980
Granted
Sep 13, 2016
Kind
B2
Abstract

A collision prediction apparatus includes a sensor that obtains positional information representing a position of a target object with respect to a host vehicle and a processing device wherein the processing device calculates a movement trajectory of the target object with respect to a vehicle, based on the positional information obtained at time points by the sensor, and the processing device predicts a probability of a collision between the target object and the vehicle, based on the calculated movement trajectory and at least one of three parameters which includes a number of obtainment time points of the positional information used to calculate the movement trajectory, a change manner of a lateral width of the target object related to the positional information between time points, and a change manner of the positional information of the target object with respect to the vehicle in a longitudinal direction of the vehicle.

Claims (16)

1. A collision prediction apparatus, comprising:

a sensor that obtains positional information representing a position of a target object with respect to a host vehicle; and

a processor, wherein

the processor calculates a movement trajectory of the target object with respect to the host vehicle, based on the positional information obtained at a plurality of time points by the sensor, the movement trajectory being calculated using a linear approximation that excludes outliers, the outliers being positional information outside of a predetermined error range,

the processor predicts a probability of a collision between the target object and the host vehicle, based on the calculated movement trajectory and at least one of three parameters, wherein the three parameters include a number of obtainment time points of the positional information used to calculate the movement trajectory, a change in a lateral width of the target object between the time points, and a change in the positional information of the target object with respect to the host vehicle in a longitudinal direction of the host vehicle,

the at least one of three parameters including the number of obtainment time points such that the predicted probability of a collision becomes lower as the number of the obtainment time points becomes smaller,

the at least one of three parameters including the change in the lateral width of the target object between the time points such that the predicted probability of a collision becomes lower as the change in the lateral width between the time points becomes greater, and

the processor issues an instruction to control a control target device based on the predicted probability of a collision.

2. The collision prediction apparatus of claim 1 , wherein

the processor predicts the probability of a collision based on the calculated movement trajectory and the change in the positional information of the target object with respect to the host vehicle in a longitudinal direction of the host vehicle such that the predicted probability of a collision becomes lower in a state where there is an alienation between the change in the positional information and a change in the position of the host vehicle calculated based on a vehicle speed sensor.

3. The collision prediction apparatus of claim 1 , wherein

the processor predicts the probability of a collision based on the calculated movement trajectory and the change in the positional information of the target object with respect to the host vehicle in a longitudinal direction of the host vehicle such that the predicted probability of a collision becomes lower in a state where the position of the target object with respect to the host vehicle becomes farther with respect to the previous position of the target object.

4. The collision prediction apparatus of claim 1 , wherein

the processor calculates the movement trajectory by a RANSAC algorithm.

5. The collision prediction apparatus of claim 1 , wherein

the processor predicts, for respective portions of the host vehicle, the probability of a collision between the target object and the portion of the host vehicle.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Dec 19, 2014
From: NANAMI, TAKESHI
To: TOYOTA JIDOSHA KABUSHIKI KAISHA
Reel/Frame 034561/0184 →
Continuity (1)
Related Publication 20150183431A1 · Jul 2, 2015