Method for occupant classification in a motor vehicle
First Claim
Patent Images
1. A method for occupant classification in a vehicle, comprising:
- using a seat mat in a vehicle seat for the occupant classification;
generating a seat profile as a function of time, using a sensor matrix in the seat mat;
determining features on the basis of the seat profile;
performing a weight estimation for each of the features, resulting in respective weight estimates;
determining an occupant weight for the occupant classification from the respective weight estimates for the respective features;
determining a measure of quality for the occupant classification from the weight estimates and the occupant weight; and
continuing to use only occupant classifications whose measure of quality is above a predefined threshold value.
1 Assignment
0 Petitions
Accused Products
Abstract
A method for occupant classification in a vehicle is used to extract a plurality of features from a seat profile and to assign a weight estimate to each of these features. The occupant weight and a measure of quality are determined from the different weight estimates. The measure of quality determines whether the occupant weight continues to be used as the occupant classification. The weight estimation for the feature in question is carried out via a stored function. The occupant classification can be used for a restraint system. The ischial tuberosity distance and/or the seat profile size can be used as the features.
22 Citations
19 Claims
-
1. A method for occupant classification in a vehicle, comprising:
-
using a seat mat in a vehicle seat for the occupant classification;
generating a seat profile as a function of time, using a sensor matrix in the seat mat;
determining features on the basis of the seat profile;
performing a weight estimation for each of the features, resulting in respective weight estimates;
determining an occupant weight for the occupant classification from the respective weight estimates for the respective features;
determining a measure of quality for the occupant classification from the weight estimates and the occupant weight; and
continuing to use only occupant classifications whose measure of quality is above a predefined threshold value. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9)
where A is the average occupant weight, B is the first weight estimate based on the seat profile size, and C is the second weight estimate based on the ischial tuberosity distance.
-
-
9. The method of claim 1, wherein the measure of quality is a profile quality, which is determined using as follows:
- Q=(1−
[max(B, C)−
min(B,C)]/A), where Q is a value for the profile quality, A is an average occupant weight, B is a first weight estimate based on the seat profile size, and C is a second weight estimate based on the ischial tuberosity distance.
- Q=(1−
-
10. A device for occupant classification in a vehicle, comprising:
-
a seat mat situated in a vehicle seat for the occupant classification, the seat mat having a sensor matrix for generating a seat profile as a function of time; and
a processor for performing the following;
determining features on the basis of the seat profile, performing a weight estimation for each of the features, resulting in respective weight estimates, determining an occupant weight for the occupant classification from the respective weight estimates for the respective features, determining a measure of quality for the occupant classification from the weight estimates and the occupant weight, and continuing to use only occupant classifications whose measure of quality is above a predefined threshold value. - View Dependent Claims (11, 12, 13, 14, 15, 16, 17, 18, 19)
where A is the average occupant weight, B is the first weight estimate based on the seat profile size, and C is the second weight estimate based on the ischial tuberosity distance.
-
Specification