The present application discloses a robotic control system, and a method and a computer system for controlling a robot. The robotic control system includes a memory and one or more processors coupled to the memory. The memory is configured to store configured to store a model of a robot having a plurality of axes of control including at least a linear axis and one or more rotational axes. The one or more processors are configured to use the model to control the robot to perform a task, including by sending to the robot a set of control signals to cause the robot to move with respect to two or more of said axes of control including at least the linear axis.

Packing store items in a shopping cart using a robotic device can include determining packing preferences of a user. A product and at least one attribute of the product can be identified in response to the user placing the product in the shopping cart. In response to the user placing the product among other products in the shopping cart, the product and other products in the shopping cart can be arranged based on the packing preference and at least one attribute of each product. Packing of products can be initiated in response to detecting completion of shopping by the user.

A robotic assistance device includes a robotic arm configured assist a human being and one or more sensors configured to receive input data. The input data includes information related to one or more activities being performed by the human being. The robotic assistance device further includes a processor operatively connected to the one or more sensors and the first robotic arm. The processor is configured to process data received from the sensor, and to determine, based on the processed data, that a task should be performed by the robotic assistance device based at least in part on the activities being performed by the human being. The processor is further configured cause the operative piece to perform the task.

A method for training a machine learning model for recognizing an object topology of an object from an image of the object. The method includes obtaining a 3D model of the object, wherein the 3D model comprises a mesh of vertices connected by edges, wherein each edge has a weight which specifies proximity of two vertices connected by the edge in the object; determining a descriptor for each vertex of the mesh by searching descriptors for the vertices which minimize the sum, over pairs of connected vertices, of distances between the descriptors of the pair of vertices weighted by the weight of the edge between the pair of vertices; generating training data image pairs, wherein each training data image pair comprises a training input image showing the object and a target image; and training the machine learning model by supervised learning using the training data image pairs as training data.

The present application discloses a mechanism to adjust backlash in a rack and pinion powertrain assembly. The mechanism to adjust backlash includes a mounting frame having an opening defined therein to receive an operative end of a drive assembly, a shoulder fastener positioned through a first complementary set of holes at a first end of a mounting flange to movably couple the mounting flange to the mounting frame, the fastener being fastened in a manner such that the mounting flange and drive assembly have freedom to pivot about a longitudinal axis of the first complementary set of holes, and an adjustable length coupling device having a first end coupled mechanically to the mounting plate and a second end coupled mechanically to the mounting flange at a location substantially opposite the first end of the mounting flange.

In a robot controller, return of a position (including posture) of a robot arm to an origin thereof is controlled along a searched return path, when there occurs an abnormality against the robot. It is determined whether there is a trajectory portion regarded as being a straight line on a motion trajectory of the arm, based on the operation log data recorded in a recording unit when it is determined that the arm should be returned to the origin. Operation log data of a node are decimated (i.e., downsampled), which are present between two nodes which are at both ends of a trajectory portion regarded as being a straight line when the trajectory portion is regarded as being the straight line. Then, the return path is calculated, along which the arm returns to the origin, based on remaining operation log data.

Systems (100) and methods (900) for controlling movement of an articulating arm having a plurality of joints. The methods comprise: receiving, by the controller, a command to perform a task by the articulating arm; ranking movements of the joints based on how much each said joint needs to move at a first time in order to follow the command; selecting a first subset of joints with top-ranked movements from the plurality of joints, where the subset of joints comprises less than a total number of joints contained in the plurality of joints; and causing only the joints of the first subset to move during a first timeslot of a plurality of timeslots.

A method for controlling a robot using control parameter values from a non-Euclidean control parameter space. The method includes performing a Bayesian optimization of an objective function representing a desired control objective of the robot over the control parameter space, wherein evaluation points of the objective function are determined by searching an optimum of an acquisition function in an iterative search. In each iteration, the following are performed: updating a candidate evaluation point using a search direction in the tangent space of the parameter space at the candidate evaluation point, mapping the updated candidate evaluation point from the tangent space to the parameter space, and using the mapped updated candidate evaluation point as evaluation point for a next iteration until a stop criterion is fulfilled, and controlling the robot in accordance with a control parameter value found in the Bayesian optimization.

The present invention relates to methods, devices and systems for associating consumable data with an assay consumable used in a biological assay. Provided are assay systems and associated consumables, wherein the assay system adjusts one or more steps of an assay protocol based on consumable data specific for that consumable. Various types of consumable data are described, as well as methods of using such data in the conduct of an assay by an assay system. The present invention also relates to consumables (e.g., kits and reagent containers), software, data deployable bundles, computer-readable media, loading carts, instruments, systems, and methods, for performing automated biological assays.

A surgical robot comprising a surgical robot arm and a surgical robot arm controller. The surgical robot arm comprises a set of joints and a joint controller. The joint controller is configured to drive a joint of the set of joints. The surgical robot arm controller comprises a processor and watchdog circuitry. The processor is configured to send joint driving signals to the joint controller on a communication link. The watchdog circuitry is configured to: receive sequence values from the processor; determine whether each received sequence value matches a next expected value of a predetermined sequence; and if the received sequence value does not match the next expected value of the predetermined sequence, disable the communication link between the processor and the joint controller.