Methods and devices for a vehicle

Patent metadata

Field	Value
Publication number	US-12545158-B2
Application number	US-202117469932-A
Country	US
Kind code	B2
Filing date	Sep 9, 2021
Priority date	Sep 9, 2021
Publication date	Feb 10, 2026
Grant date	Feb 10, 2026

Abstract

Devices and methods for a vehicle are provided in this disclosure. A device for controlling an active seat of a vehicle may include a processor and a memory. The memory may be configured to store a transfer function. The processor may be configured to predict an acceleration of the active seat of the vehicle based on a first sensor data and the transfer function. The first sensor data may include information indicating an acceleration of a vibration source for the vehicle. The processor may be further configured to generate a control signal to control a movement of the active seat at a first instance of time based on the predicted acceleration. Furthermore, the processor may be configured to adjust the transfer function based on a second sensor data including information indicating a detected acceleration of the active seat at the first instance of time.

First claim

What is claimed is:

1 . A device comprising:
a memory configured to store a transfer function; a processor configured to:
predict an acceleration of an active seat of a vehicle based on a first sensor data and the transfer function, wherein the first sensor data comprises information indicating an acceleration of a vibration source for the vehicle;
generate a control signal to control an actuation of the active seat relative to a chassis of the vehicle at a first instance of time, wherein the control signal is generated by selectively operating in one of a first control mode or a second control mode,
wherein the selection is based on a comparison of a detected acceleration of the active seat at a second instance of time, temporally before the first instance of time, with a predefined threshold,
wherein, in the first control mode that is selected when the detected acceleration is below the predefined threshold, the control signal is generated based on the predicted acceleration and the detected acceleration, and
wherein, in the second control mode that is selected when the detected acceleration is above the predefined threshold, the control signal is generated based on the detected acceleration without being based on the predicted acceleration;
control the actuation of the active seat according to the control signal;
adjust the transfer function based on a second sensor data comprising information indicating a detected acceleration of the active seat at the first instance of time.

2 . The device of claim 1 ,
wherein the actuation of the active seat comprises at least one of the following: a spatial movement of the active seat, an angular movement of the active seat, a spatial movement of a part of the active seat, an angular movement of a part of the active seat, a change of a shape of the active seat, and/or a change of at least one property of the active seat.

3 . The device of claim 1 ,
wherein the processor is further configured to adjust the transfer function based on the predicted acceleration of the active seat and the second sensor data.

4 . The device of claim 1 ,
wherein the first sensor data comprises information indicating at least one of an angle of a steering wheel of the vehicle, a throttle of the vehicle, or a brake of the vehicle comprising a plurality of sensor data items received from a plurality of inertial measurement unit (IMU) sensors, wherein a first inertial measurement unit (IMU) sensor is configured to detect an acceleration of an engine, wherein a second inertial measurement unit (IMU) sensor is configured to detect an acceleration of a wheel of the vehicle.

5 . The device of claim 1 ,
wherein the memory is further configured to store a plurality of model parameters, wherein the processor is further configured to predict the acceleration of the active seat of the vehicle with the transfer function by applying a neural network model to the first sensor data according to the plurality of model parameters, wherein the neural network model is configured to use a cost function that is configured to adjust at least one of the plurality of model parameters by penalizing a difference between the predicted acceleration and the detected acceleration of the active seat at the first instance of time.

6 . The device of claim 5 ,
wherein the neural network model is configured to use the cost function to keep the difference between the predicted acceleration and the detected acceleration of the active seat below a predefined human sensitivity threshold, wherein the cost function is configured to be equal to |{umlaut over (X)} s −{umlaut over (X)} p | 2 2 +∥{umlaut over (X)} s −{umlaut over (X)} p |−{right arrow over (ω)}| 2 2 , wherein {umlaut over (X)} s is the detected acceleration, X p is the predicted acceleration, and {right arrow over (ω)} is a predefined threshold.

7 . The device of claim 5 ,
wherein the neural network model is configured to use the cost function to keep the difference between the predicted acceleration and the detected acceleration of the active seat below a predefined human sensitivity threshold.

8 . The device of claim 7 ,
wherein that the neural network model is configured to keep the difference between the predicted acceleration and the detected acceleration of the active seat below 0.1 m/s2 for each axis.

9 . The device of claim 1 ,
wherein the processor comprises a feedforward proportional derivative controlling block configured to control the actuation of the active seat relative to a chassis of the vehicle at the first instance of time based on the predicted acceleration, wherein the feedforward proportional derivative controlling block is configured to generate a first control signal based on the detected acceleration at a second instance of time, wherein the second instance of time is temporally before the first instance of time.

10 . The device of claim 9 ,
wherein the processor comprises a feedback proportional derivative controlling block that is configured to generate a second control signal based on the detected acceleration at the second instance of time, wherein the processor is further configured to generate the control signal based on the first control signal and the second control signal to control the actuation of the active seat relative to the chassis of the vehicle at the first instance of time based on the predicted acceleration.

11 . The device of claim 10 ,
wherein the processor is further configured to:
generate the control signal based on the first control signal and the second control signal, in case the detected acceleration of the active seat for the second instance of time is below a predefined threshold;
activate a damping feedback control mode, in case the detected acceleration of the active seat for the second instance of time is above a predefined threshold;
generate the control signal for the first instance of time only based on the second control signal in the damping feedback control mode.

12 . The device of claim 10 ,
wherein the processor is further configured to activate a feedback off mode, in case a difference between a predicted acceleration for the second instance of time and the detected acceleration for the second instance of time is above a predefined threshold, wherein the processor is further configured to indicate that the active seat is not to be moved.

13 . The device of claim 10 ,
wherein the predefined threshold comprises 0.1 m/s2 for at least one axis.

14 . The device of claim 1 ,
wherein the processor is further configured to:
apply a reinforcement learning model according to a plurality of model parameters in the memory;
generate a reward metric based on the detected acceleration in response to the predicted acceleration; and
adjust the plurality of model parameters based on the reward metric.

15 . The device of claim 1 ,
wherein the information indicating the acceleration of the vibration source comprises information of a translational envelope comprising an x-axis, a y-axis, and a z-axis.

16 . The device of claim 1 ,
wherein the information indicating the acceleration of the vibration source comprises information of a rotational envelope comprising a rotation around an x-axis, a rotation around a y-axis, and a rotation around a z-axis.

17 . The device of claim 1 ,
wherein the memory is further configured to store a plurality of model parameters, wherein the processor is further configured to predict the acceleration of the active seat of the vehicle with the transfer fun

Assignees

• Intel Corp

Inventors

• Alvarez Ignacio J
• Alyuz Civitci Nese
• Elli Maria Soledad
• Felip Leon Javier
• Geissler Florian
• Gonzalez Aguirre David Israel
• Kose Cihangir Neslihan
• Paulitsch Michael
• Rosales Rafael
• Turek Javier S

Classifications

• B60N2230/20

  Wireless data transmission · CPC title

• B60N2230/10

  Wired data transmission · CPC title

• B60N2220/30

  using a neural network · CPC title

• B60N2220/20

  using a deterministic algorithm · CPC title

• B60N2230/30

  Signal processing of sensor data · CPC title