Matched filter for vehicle crash discrimination

• US 5,559,699 A
• Filed: 08/26/1994
• Issued: 09/24/1996
• Est. Priority Date: 08/26/1994
• Status: Expired due to Term

• Current Assignee: Automotive Systems Laboratory, Incorporated (Ningbo Junsheng Investment Group Co., Ltd.)
• Original Assignee: Automotive Systems Laboratory, Incorporated (Ningbo Junsheng Investment Group Co., Ltd.)
• Inventors: Gioutsos, Tony
• Primary Examiner(s): NOT, DEFINED
• Assistant Examiner(s): NOT, DEFINED
• Application Number:
  US08/296,524
• Time in Patent Office: 760 Days
• Field of Search: 364/424.05, 364/424.01, 340/436, 340/437, 340/438, 280/735, 180/274, 180/282, 307/10.1
• US Class Current: 701/45
• CPC Class Codes: B60R 21/0132 responsive to vehicle motion parameters , e.g. to vehicle longitudinal or transversal deceleration or speed value
• CPC Class Codes: B60R 21/01336 using filtering

Abstract

A matched filter (14) for vehicle crash discrimination system (10) utilizes at least one set of n scaling factors stored in a memory (24) to combine (22) with a set of n time sequenced acceleration data samples. The values of the n scaling factors are selected to effect a desired vehicle crash discrimination strategy. The matched filter optimizes system design flexibility and reconfigurability by allowing different strategies to be implemented simply by selecting the appropriate scaling factors stored in the memory.

Claims

1. A matched filter for processing data representative of vehicle acceleration to produce a discrimination output value suitable for threshold comparison to determine whether to deploy an occupant safety restraint device, said filter comprising:

an input means for organizing the acceleration data into n time sequential data samples ranging from a current sample to an n^th oldest sample;

a memory means for storing at least one set of n predetermined scaling factors;

means for combining each of the n data samples with a predefined one of the n scaling factors, wherein the respective values of said n scaling factors are predetermined to effect a predetermined vehicle crash discrimination strategy; and

means for combining the n scaled data samples to produce the discrimination output value.

2. The matched filter of claim 1 wherein the respective values of said at least one set of n scaling factors are selected to exponentially reduce the oldest acceleration data samples to provide a discrimination output value which is damped with respect to past vehicle acceleration.

3. The matched filter of claim 1 wherein said at least one set of n scaling factors comprises a first and second set of n scaling factors, said first set of scaling factors selected to provide a discrimination output value representative of vehicle velocity, and said second set selected to provide a discrimination output value representative of a jerk value.