SYSTEM AND METHOD FOR DETECTING PRESENCE OF A HUMAN IN A VEHICLE

• US 20060089753A1
• Filed: 10/25/2004
• Published: 04/27/2006
• Est. Priority Date: 10/25/2004
• Status: Active Grant

• Current Assignee: Ford Global Technologies, LLC (Ford Motor Company)
• Original Assignee: Ford Global Technologies, LLC (Ford Motor Company)
• Inventors: Smith, Greg, Prakah-Asante, Kwaku, Ebenstein, Sam, Rodin, Yelena, Feldkamp, Lee, Eagen, Charles
• Granted Patent: US 7,353,088 B2
• Time in Patent Office: Days
• US Class Current: 701/1
• CPC Class Codes: B60R 25/1004 Alarm systems characterised by the type of sensor, e.g. current sensing means using door lock sensors B60R25/1001
• CPC Class Codes: G08B 13/1663 using seismic sensing means

Abstract

A system for detecting the presence of a human in a vehicle is provided. The system includes a vibration sensor that is configured to detect vibrations of the vehicle, and to output signals related to the sensed vibrations. A processor is configured to receive the signals output from the vibration sensor. The processor also operates a neural network that has a plurality of nodes, at least some of which are recurrent. The use of the recurrent nodes allows the output of a recurrent node to be fed back into itself, or another node. In addition, the output that is fed back can be combined with other inputs entering the node. In this way, the neural network can quickly learn to distinguish between various conditions, including an occupied state and an unoccupied state of the vehicle. The neural network provides an output indicating whether the vehicle is occupied.

Claims

1. A system for detecting the presence of a human in a vehicle, the system comprising:

a vibration sensor configured to detect vibrations of the vehicle and to output signals related to the sensed vibrations;

a processor configured to receive the signals output from the vibration sensor; and

a neural network run by the processor and having plurality of nodes, at least one of the nodes being a recurrent node, thereby facilitating operation of the neural network using a time delay between an output from a recurrent node and an input, of the recurrent node output, into another of the nodes, and using a time delay between the recurrent node output and an input, of the recurrent node output, back into the recurrent node, the neural network being configured to provide at least one output value indicating that a human is present in the vehicle and at least one output value indicating that a human is not present in the vehicle.

2. The system of claim 1, wherein the neural network can be operated such that the recurrent node output can be directly input back into the recurrent node, and

wherein the neural network can be operated such that the recurrent node output can be input into another node.

3. The system of claim 1, wherein the output values from the neural network are based on a magnitude of the signals received by the processor.

4. The system of claim 1, wherein the output values from the neural network are based on a shape of the signals received by the processor.

5. The system of claim 1, wherein the signals output from the vibration sensor are analog signals, the system further comprising:

an amplifier in communication with the vibration sensor and configured to amplify the analog signals output by the vibration sensor; and

a convertor for converting the amplified signals from an analog format to a digital format before the signals are received by the processor.

6. The system of claim 5, further comprising a filter for filtering out signals outside a predetermined frequency range, thereby limiting the frequency range of the signals received by the processor.

7. The system of claim 1, wherein the neural network is trained to distinguish between a human being present in the vehicle and a human not being present in the vehicle by collecting data from the vibration sensor over a plurality of vehicle conditions.

8. The system of claim 7, wherein the neural network is further configured to provide at least one output value indicating that the neural network is uncertain as to the presence of a human in the vehicle.

9. The system of claim 1, wherein each recurrent node is configured to receive a plurality of inputs and to provide a single output to another node based on the inputs.

10. The system of claim 9, wherein the inputs to each recurrent node are mathematically combined to generate the single output.

11. A vehicle including a system for detecting the presence of a human in the vehicle, the vehicle comprising:

a vibration sensor disposed on a portion of the vehicle for detecting vibrations of the vehicle and for outputting signals related to the sensed vibrations;

a processor configured to receive the signals output from the vibration sensor; and

a neural network run by the processor and having plurality of nodes, at least one of the nodes being a recurrent node, thereby facilitating operation of the neural network using a time delay between an output from a recurrent node and an input, of the recurrent node output, into another of the nodes, and using a time delay between the recurrent node output and an input, of the recurrent node output, back into the recurrent node, the neural network being configured to provide at least one output value indicating that a human is present in the vehicle and at least one output value indicating that a human is not present in the vehicle.

12. The vehicle of claim 11, wherein the neural network can be operated such that the recurrent node output can be directly input back into the recurrent node, and

wherein the neural network can be operated such that the recurrent node output can be input into another node.

13. The vehicle of claim 11, further comprising a suspension system, and wherein the vibration sensor is disposed on the portion of the vehicle such that the vibration sensor detects vibrations of suspension system.

14. The vehicle of claim 11, wherein the outputs from the neural network are based on at least one of a magnitude of the signals received by the processor and a shape of the signals received by the processor.

15. The vehicle of claim 11, wherein the neural network is trained to distinguish between a human being present in the vehicle and a human not being present in the vehicle by collecting data from the vibration sensor over a plurality of vehicle conditions.

16. The vehicle of claim 15, wherein the neural network is further configured to provide at least one output value indicating that the neural network is uncertain as to the presence of a human in the vehicle.

17. The vehicle of claim 11, wherein each recurrent node is configured to receive a plurality of inputs and to provide a single output to another node based on the inputs.

18. The vehicle of claim 17, wherein the inputs to each recurrent node is mathematically combined to generate the single output.

19. A method for detecting the presence of a human in a vehicle, the method comprising:

sensing vibrations in the vehicle;

outputting signals related to the sensed vibrations;

processing the signals using a neural network having a plurality of nodes, at least one of the nodes being a recurrent node, thereby facilitating operation of the neural network using a time delay between an output from a recurrent node and an input, of the recurrent node output, into another of the nodes, and using a time delay between the recurrent node output and an input, of the recurrent node output, back into the recurrent node; and

outputting a signal from the neural network indicating whether a human is present in the vehicle.

20. The method of claim 19, wherein the signal output from the neural network indicating whether a human is present in the vehicle is based on at least one of a magnitude of the sensed vibrations and a shape of the sensed vibrations.

21. The method of claim 19, further comprising training the neural network to distinguish between a human being present in the vehicle and a human not being present in the vehicle, the training including sensing vibrations over a plurality of vehicle conditions.

22. The method of claim 20, further comprising outputting a signal from the neural network indicating that the neural network is uncertain as to the presence of a human in the vehicle.