Monitoring system for a vehicle cabin

A vehicle, comprising:

• a vehicle cabin, and
• a monitoring system for the vehicle cabin, the monitoring system comprising:
• a first sensor unit,
• a second sensor unit, and
• a control unit,
• the first sensor unit being arranged in a longitudinal direction of the vehicle and being configured to generate image data of the vehicle cabin,
• the second sensor unit being arranged perpendicular to a driving direction of the vehicle and being configured to generate non-image data of the vehicle cabin,
• the control unit being coupled to a plurality of subsystems and configured to initiate scanning the vehicle cabin by the first sensor unit and/or the second sensor unit based on a command for actuation of a subsystem of the plurality of subsystems,
• the control unit being further configured to collect and fuse a determined level of the image data and the non-image data after scanning is initiated and determine based thereon whether an obstacle is in the vehicle cabin,
• the control unit being further configured to determine what level of the image data and the non-image data to fuse with respect to a degree of processing that has been done to the image data and the non-image data, being raw data, feature extraction data, or inference data, based on training for sensor type and detection result, and the control unit being further configured to limit the actuation of the subsystem if the obstacle is disruptive for the actuation of the subsystem.

A method for monitoring a vehicle cabin, the method comprising:

• generating image data of the vehicle cabin by a first sensor unit being arranged in a longitudinal direction of a vehicle,
• generating non-image data of the vehicle cabin by a second sensor unit being arranged perpendicular to a driving direction of the vehicle,
• initiating scanning the vehicle cabin by the first sensor unit and/or the second sensor unit at the direction of a control unit coupled to a plurality of subsystems based on a command for actuation of a subsystem of the plurality of subsystems,
• determining what level of the image data and the non-image data to fuse with respect to a degree of processing that has been done to the image data and the non-image data, being raw data, feature extraction data, or inference data, based on training for sensor type and detection result,
• collecting and fusing the determined level of the image data and the non-image data after scanning is initiated and determining based thereon whether an obstacle is in the vehicle cabin, and
• limiting the actuation of the subsystem if the obstacle is disruptive for the actuation of the subsystem.