Abstract: A gripper includes a lead screw, a first gripping jaw and a second gripping jaw. The lead screw includes a spiral portion having at least one first spiral track and at least one second spiral track with opposite helical directions. A portion of a coverage of the first spiral track overlaps a portion of a coverage of the second spiral track along a length of the spiral portion. The first gripping jaw has a first pin extending into the first spiral track, and the second gripping jaw has a second pin extending into the second spiral track. When the spiral portion rotates, the first spiral track drives the first pin to allow a first linear movement of the first gripping jaw, and the second spiral track drives the second pin to allow a second linear movement of the second gripping jaw opposite to the first linear movement.

Abstract: The present disclosure relates to a brake mechanism, a joint actuator and a robot. The brake mechanism includes a friction member configured to be fixed to a rotor of the motor, a brake member abutting against one side of the friction member, a pushing member abutting against the other side of the friction member and configured to provide an adjustable pushing force to the brake member, a locking mechanism configured to prevent the brake member from rotating according to a brake command.

Abstract: A gripper with at least one gripping jaw is provided. The at least one gripping jaw includes a base, a gripping portion configured to contact an object, a flexible member connecting the base and the gripping portion, and a sensor assembly located between the base and the gripping portion. The flexible member is configured to enable the gripping portion to deflect with respect to the base when the gripping portion is subjected to a force in a first direction from the object. The sensor assembly is configured to generate a signal in response to the deflection of the gripping portion.

Abstract: A method of processing data for a target model, an electronic device, and a storage medium, which relate to a field of deep learning. The method of processing data for a target model includes: acquiring the target model, wherein the target model includes at least one network layer, and each network layer of the at least one network layer includes a plurality of model parameters; for a target network layer in the at least one network layer, dividing the plurality of model parameters of the target network layer into a plurality of groups; and adjusting a data type of the model parameters in the plurality of groups from a first data type to a second data type respectively, so as to obtain a processed target model.

Abstract: A gripper includes a lead screw, a first gripping jaw and a second gripping jaw. The lead screw includes a spiral portion having at least one first spiral track and at least one second spiral track with opposite helical directions. A portion of a coverage of the first spiral track overlaps a portion of a coverage of the second spiral track along a length of the spiral portion. The first gripping jaw has a first pin extending into the first spiral track, and the second gripping jaw has a second pin extending into the second spiral track. When the spiral portion rotates, the first spiral track drives the first pin to allow a first linear movement of the first gripping jaw, and the second spiral track drives the second pin to allow a second linear movement of the second gripping jaw opposite to the first linear movement.

Abstract: The present disclosure relates to a gripper and a robot. The gripper according to the present disclosure includes at least three guiding rails arranged head to tail in sequence, at least three gripping fingers respectively disposed on the at least three guiding rails and a driving mechanism. A first end of each of the gripping fingers is slidably connected to the corresponding guiding rail and a second end of each of the gripping fingers is for contacting the object to be gripped. The driving mechanism drives each of the gripping fingers to move along the corresponding guiding rail, so that the second end of each of the gripping fingers moves toward or away from a gripping center of the gripper.

Abstract: The present disclosure relates to a brake mechanism, a joint actuator and a robot. The brake mechanism includes a friction member configured to be fixed to a rotor of the motor, a brake member abutting against one side of the friction member, a pushing member abutting against the other side of the friction member and configured to provide an adjustable pushing force to the brake member, a locking mechanism configured to prevent the brake member from rotating according to a brake command.

Abstract: The present disclosure relates to a brake mechanism, a joint actuator and a robot. The brake mechanism includes a friction member configured to be fixed to a rotor of the motor, a brake member abutting against one side of the friction member, a pushing member abutting against the other side of the friction member and configured to provide an adjustable pushing force to the brake member, a locking mechanism configured to prevent the brake member from rotating according to a brake command.

Abstract: The present disclosure relates to a brake mechanism, a joint actuator and a robot. The brake mechanism includes a friction member configured to be fixed to a rotor of the motor, a brake member abutting against one side of the friction member, a pushing member abutting against the other side of the friction member and configured to provide an adjustable pushing force to the brake member, a locking mechanism configured to prevent the brake member from rotating according to a brake command.

Abstract: A method for estimating a blur kernel size, the method including: (1) pre-processing a blurred image, to obtain an image, so that a size of the image satisfies an image input size of a multi-class convolutional neural network (CNN); (2) inputting the image into a multi-class CNN with completed training, to obtain a blur-kernel-size probability distribution vector; and (3) comparing each element in the blur-kernel-size probability distribution vector, so that an estimated blur kernel size of the blurred image is the blur kernel size corresponding to a largest element. The invention also provides a system for estimating a blur kernel size. The system includes an image pre-processing module, a training-set synthesizing module, a multi-class CNN module, and a blur kernel size estimation module.

Abstract: A method for estimating a blur kernel size, the method including: (1) pre-processing a blurred image, to obtain an image, so that a size of the image satisfies an image input size of a multi-class convolutional neural network (CNN); (2) inputting the image into a multi-class CNN with completed training, to obtain a blur-kernel-size probability distribution vector; and (3) comparing each element in the blur-kernel-size probability distribution vector, so that an estimated blur kernel size of the blurred image is the blur kernel size corresponding to a largest element. The invention also provides a system for estimating a blur kernel size. The system includes an image pre-processing module, a training-set synthesizing module, a multi-class CNN module, and a blur kernel size estimation module.

Abstract: The invention discloses a method for correcting aero-optical thermal radiation noise, comprising steps of: pretreating a degraded image to obtain a multi-scale degraded image group, conducting iteration process of obtaining an optimal solution by using a last scale estimation result as an original value of next scale estimation according to the multi-scale degraded image group, thereby facilitating original-scale bias field estimation, and restoring the degraded image according to the original-scale bias field estimated value thereby obtaining an image after aero-optical thermal radiation noise correction. The invention also discloses a system for correcting aero-optical thermal radiation noise. The invention is capable of solving problems with conventional methods, comprising poor correction effect, high complexity, and incapability in correcting the thermal radiation noise at an image level, and applicable to restoration of an image with aero-optical thermal radiation noise.

Abstract: The invention discloses a method for correcting aero-optical thermal radiation noise, comprising steps of: pretreating a degraded image to obtain a multi-scale degraded image group, conducting iteration process of obtaining an optimal solution by using a last scale estimation result as an original value of next scale estimation according to the multi-scale degraded image group, thereby facilitating original-scale bias field estimation, and restoring the degraded image according to the original-scale bias field estimated value thereby obtaining an image after aero-optical thermal radiation noise correction. The invention also discloses a system for correcting aero-optical thermal radiation noise. The invention is capable of solving problems with conventional methods, comprising poor correction effect, high complexity, and incapability in correcting the thermal radiation noise at an image level, and applicable to restoration of an image with aero-optical thermal radiation noise.