Abstract: A method classifies an IR image to be used for training a machine learning model. The method comprises: (i) providing a set of machine learning models in function of temperature; (ii) acquiring an infrared image; (iii) identifying a target object to be detected and a comparison object in the infrared image; (iv) calculating a first characteristic of the target object in the infrared image and a second characteristic of the comparison object in the infrared image; and (v) based on the first characteristic of the target object and the second characteristic of the comparison object, selecting a machine learning model among the set of machine learning models in function of temperature for the IR image.

Abstract: Aspects of the subject technology relate to a self-centering camera mount. The self-centering camera mount includes a camera mounting plate, first and second deflectors respectively arranged on a first side and a second side of the vehicle, first and second distance sensors mounted on the camera mounting plate to respectively a first distance between the first distance sensor and the first deflector and a second distance between the second distance sensor and the second deflector, and a processor that, when determining that the first distance and the second distance do not match each other, controls, based on the received first and second distances, a camera mounting plate actuator to slide the camera mounting plate until the first and second distances become equal to each other.

Abstract: Aspects of the subject technology relate to a self-centering camera mount. The self-centering camera mount includes a camera mounting plate, first and second deflectors respectively arranged on a first side and a second side of the vehicle, first and second distance sensors mounted on the camera mounting plate to respectively a first distance between the first distance sensor and the first deflector and a second distance between the second distance sensor and the second deflector, and a processor that, when determining that the first distance and the second distance do not match each other, controls, based on the received first and second distances, a camera mounting plate actuator to slide the camera mounting plate until the first and second distances become equal to each other.

Abstract: An incrementally assembled skin-constrained inflatable bellows system includes a top end cap, a bottom end cap, and a bellows attached to and separating the top end cap and the bottom end cap and configured to hold pressurized fluid between the top and bottom end caps. The system further includes a nozzle configured to allow fluid to enter and exit the bellows, and an inextensible skin attached to each of the top end cap, the bottom end cap, and to the bellows. The top end cap assumes a first position when the bellows is inflated via the nozzle and, when the top end cap is adjusted from the first position to a second position, the skin is configured to maintain the top end cap in the second position and constrains the bellows into a predefined rigid three-dimensional shape.