Abstract: Described is a control command system for generating control commands for a vehicle. The system receives a networked control system corresponding to a network of subsystems of sensors and actuators of the vehicle and data collected from sensors on the vehicle. Multiple subsystems are formed from the networked control system. A dynamic mode decomposition with control (DMDc) process is applied to each subsystem, yielding a linear approximation for each subsystem. The linear approximations are combined into a single linear approximation for the networked control system, and a linear control system approximating the networked control system is output. Control commands are generated for the vehicle based on the linear control system, which cause the vehicle to perform a vehicle operation.

Abstract: A method, system and computer program product are provided for aligning wire contacts with wire contact insertion holes of a connector to facilitate the automated insertion of the wire ends of a wire bundle assembly into the wire contact insertion holes of a connector. Methods may include: obtaining captured images from at least two image capture devices attached to an end-effector of a robot of a wire gripper of the end-effector; detecting, within at least one image from each of the at least two image capture devices, a wire contact; detecting, within at least one image from each of the at least two image capture devices, one or more insertion holes of the connector; identifying corrective movement for the robot end-effector that aligns a target hole of the one or more insertion holes of the connector with the wire connector; and causing the robot to move the end-effector according to the identified corrective movement.

Abstract: A method, system and computer program product are provided for aligning and inserting a wire contact within a target hole of a connector to facilitate the automated insertion of the wire ends of a wire bundle assembly into the wire contact insertion holes of a connector. Methods may include: obtaining captured images, from at least two image capture devices attached to an end-effector of a robot, of a wire gripper of the end-effector; causing the robot to advance the end-effector to move the wire contact within a predetermined distance of the connector; causing the robot to advance the end-effector to move the wire contact toward the connector a predetermined additional amount more; and identifying, based on movement of the wire contact the predetermined additional amount more, if alignment is correct from force feedback at the wire gripper.

Abstract: Described is a system for co-adaptation of robot control to human biomechanics. During operation, the system receives joint angle and joint velocity of a human and co-robot and generates estimated internal states of the human. A task space motion plan is then generated for the co-robot based on a specified cooperative task and estimated internal states and joint angle and joint velocity of the human. Joint torque commands are then generated based on the task space motion plan and joint angle and joint velocity of the human and co-robot. Motion of the co-robot is then controlled, such as causing the co-robot to actuate one or more actuators to move based on the joint torque commands.

Abstract: Described is a system for action recognition through application of deep embedded clustering. For each image frame of an input video, the system computes skeletal joint-based pose features representing an action of a human in the image frame. Non-linear mapping of the pose features into an embedded action space is performed. Temporal classification of the action is performed and a set of categorical gesture-based labels is obtained. The set of categorical gesture-based labels is used to control movement of a machine.

Abstract: Described is a system for detecting backdoor attacks in deep convolutional neural networks (CNNs). The system compiles specifications of a pretrained CNN into an executable model, resulting in a compiled model. A set of Universal Litmus Patterns (ULPs) are fed through the compiled model, resulting in a set of model outputs. The set of model outputs are classified and used to determine presence of a backdoor attack in the pretrained CNN. The system performs a response based on the presence of the backdoor attack.

Abstract: Described is a system for detecting change of context in a video stream on an autonomous platform. The system extracts salient patches from image frames in the video stream. Each salient patch is translated to a concept vector. A recurrent neural network is enervated with the concept vector, resulting in activations of the recurrent neural network. The activations are classified, and the classified activations are mapped onto context classes. A change in context class is detected in the image frames, and the system causes the autonomous platform to perform an automatic operation to adapt to the change of context class.

Abstract: What is illustrated and described is a pump assembly (1) for a pump, particularly for a miniature pump, with an electric motor (2) and with a housing unit (3), wherein the electric motor (2) has a motor shaft (5) passing freely through a motor housing of the electric motor (2) and the motor housing is connected to the housing unit (3), and wherein the motor shaft (5) is supported in the motor housing with an axial shaft play. According to the invention, at least one stabilizing element (6, 13, 14, 16) arranged outside of the motor housing is provided in order to reduce or eliminate the axial play of the motor shaft (5), with an axial compressive force or axial tensile force acting on the motor shaft (3) being applied by the stabilizing element (6, 13, 14, 16) to the motor shaft (5).

Abstract: An apparatus to generate a model of a surface of an object includes a data set pre-aligner configured to receive multiple sets of surface data that correspond to respective portions of a surface of an object and that include three-dimensional (3D) points. The data set pre-aligner is also configured to perform a pre-alignment of overlapping sets to generate pre-aligned sets, including performing a rotation operation on a second set of the surface data, relative to a first set of the surface data that overlaps the second set, to apply a rotation amount that is selected from among multiple discrete rotation amounts and based on a similarity metric. The apparatus includes a data set aligner configured to perform an iterative alignment of the pre-aligned sets to generate aligned sets. The apparatus also includes a 3D model generator configured to combine the aligned sets to generate a 3D model of the object.

Abstract: Described is a system for understanding machine-learning decisions. In an unsupervised learning phase, the system extracts, from input data, concepts represented by a machine-learning (ML) model in an unsupervised manner by clustering patterns of activity of latent variables of the concepts, where the latent variables are hidden variables of the ML model. The extracted concepts are organized into a concept network by learning functional semantics among the extracted concepts. In an operational phase, a subnetwork of the concept network is generated. Nodes of the subnetwork are displayed as a set of visual images that are annotated by weights and labels, and the ML model per the weights and labels.

Abstract: An apparatus to generate a model of a surface of an object includes a data set pre-aligner configured to receive multiple sets of surface data that correspond to respective portions of a surface of an object and that include three-dimensional (3D) points. The data set pre-aligner is also configured to perform a pre-alignment of overlapping sets to generate pre-aligned sets, including performing a rotation operation on a second set of the surface data, relative to a first set of the surface data that overlaps the second set, to apply a rotation amount that is selected from among multiple discrete rotation amounts and based on a similarity metric. The apparatus includes a data set aligner configured to perform an iterative alignment of the pre-aligned sets to generate aligned sets. The apparatus also includes a 3D model generator configured to combine the aligned sets to generate a 3D model of the object.

Abstract: Described is a system for discriminant localization of objects. During operation, the system causes one or more processors to perform an operation of identifying an object in an image using a multi-layer network. Features of the object are derived from the activations of two or more layers of the multi-layer network. The image is then classified to contain one or more object classes, and the desired object class is localized. A device can then be controlled based on localization of the object in the image. For example, a robotic arm can be controlled to reach for the object.

Abstract: Described is a system for decreasing the frequency of large cascading failures in a transmission network. Based on sensors distributed throughout the transmission network, the system determines if a cascading failure is present in a transmission network. Following determination of the cascading failure, the system activates at least one switch of a plurality of switches distributed in the transmission network in order to switch transmission lines, thereby altering connectivity in the transmission network.

Abstract: Described is a system for detecting moving objects. During operation, the system obtains ego-motion velocity data of a moving platform and generates a predicted image of a scene proximate the moving platform by projecting three-dimensional (3D) data into an image plane based on pixel values of the scene. A contrast image is generated based on a difference between the predicted image and an actual image taken at a next step in time. Next, an actionable prediction map is then generated based on the contrast mage. Finally, one or more moving objects may be detected based on the actionable prediction map.

Abstract: Described is a system for classifying novel objects in imagery. In operation, the system extracts salient patches from a plurality of unannotated images using a multi-layer network. Activations of the multi-layer network are clustered into key attribute, with the key attributes being displayed to a user on a display, thereby prompting the user to annotate the key attributes with class label. An attribute database is then generated based on user prompted annotations of the key attributes. A test image can then be passed through the system, allowing the system to classify at least one object in the test image by identifying an object class in the attribute database. Finally, a device can be caused to operate or maneuver based on the classification of the at least one object in the test image.

Abstract: Described is a system for self-organized critical image segmentation. During operation, the system generates a delta pattern from a self-organized critical process. An initial test pattern is then altered based on the delta pattern to generate a new test pattern. The new test pattern is a mask identifying distinct regions in an image. A new energy score is then generated of the new test pattern. The operations of generating the delta pattern and altering the initial test pattern are then repeated until an energy score of the new test pattern is less than an energy score of the initial test pattern. At that point, the initial test pattern is replaced with the new test pattern. Finally, the process is repeated until a termination condition is reached, at which point the new test pattern provides the image segmentation by dividing the image into distinct regions, including a foreground and background.

Abstract: A method for performing an operation on an object includes capturing a plurality of images of the object. Each image is a different view of the object. The method also includes generating a sparse 3D point cloud from the plurality of images. The sparse 3D point cloud defines a 3D model of the object. The sparse 3D point cloud includes a multiplicity of missing points that each correspond to a hole in the 3D model that renders the 3D model unusable for performing the operation on the object. The method additionally includes performing curvature-based upsampling to generate a denser 3D point cloud. The denser 3D point cloud includes a plurality of filled missing points. The missing points are filled from performance of the curvature-based upsampling. The denser 3D point cloud defines a dense 3D model that is useable for performing the operation on the object.

Abstract: Described is a system and method for the classification of agents based on agent movement patterns. In operation, the system receives position data of a moving agent from a camera or sensor. Motion data of the moving agent is then extracted and used to generate a predicted future motion of the moving agent using a set of pre-calculated Echo State Networks (ESN). Each ESN represents an agent classification and generates a predicted future motion. A prediction error is generated for each ESN by comparing the predicted future motion for each ESN with actual motion data. Finally, the agent is classified based on the ESN having the smallest prediction error.

Abstract: Described is a system for decoding recorded signals into movement commands of a prosthetic device. Using a biomechanical model and physical action data, biological signal data is related to kinetic data. The physical action data can include position, joint angle, speed, and acceleration of a part of a limb. The biological signal data can include recorded neural signals and recorded muscle signals. The kinetic data can include force, power, torque, and stress. Based on the relationship between the biological signal data and the kinetic data, control commands are generated to achieve an intended position and/or movement of a prosthesis.

Abstract: Described is a system for detecting change of context in a video stream on an autonomous platform. The system extracts salient patches from image frames in the video stream. Each salient patch is translated to a concept vector. A recurrent neural network is enervated with the concept vector, resulting in activations of the recurrent neural network. The activations are classified, and the classified activations are mapped onto context classes. A change in context class is detected in the image frames, and the system causes the autonomous platform to perform an automatic operation to adapt to the change of context class.