This application relates to computing circuitry, and in particular to analogue computing circuitry suitable for neuromorphic computing. An analogue computation unit for processing data is supplied with a first voltage from a voltage regulator which is operable in a sequence of phases to cyclically regulate the first voltage. A controller is configured to control operation of the voltage regulator and/or the analogue computation unit, such that the analogue computation unit processes data during a plurality of compute periods that avoid times at which the voltage regulator undergoes a phase transition which is one of a predefined set of phase transitions between defined phases in said sequence of phases. This avoids performing computation operations during a phase transition of the voltage regulator that could result in a transient or disturbance in the first voltage, which could adversely affect the computing.

A method for moving a load with a crane in a collision-free manner in a space having at least one obstacle includes providing a position of the obstacle, providing at least one safe state variable of the load, determining from the safe state variable a safety zone surrounding the load, and dynamically monitoring the safety zone in relation to the position of the obstacle.

According to one embodiment, a calculating device includes a nonlinear oscillator. The nonlinear oscillator includes a circuit part including a first Josephson junction and a second Josephson junction, and a conductive member including a first terminal. An electrical signal is input to the first terminal. The electrical signal includes a first signal in a first operation. The first signal includes a first frequency component having a first frequency, and a second frequency component having a second frequency. The first frequency is 2 times an oscillation frequency of the nonlinear oscillator. An absolute value of a difference between the first frequency and the second frequency is not more than 0.3 times the first frequency.

According to one embodiment, a calculating device includes a nonlinear oscillator. The nonlinear oscillator includes a circuit part including a first Josephson junction and a second Josephson junction, and a conductive member including a first terminal. An electrical signal is input to the first terminal. The electrical signal includes a first signal in a first operation. The first signal includes a first frequency component having a first frequency, and a second frequency component having a second frequency. The first frequency is 2 times an oscillation frequency of the nonlinear oscillator. An absolute value of a difference between the first frequency and the second frequency is not more than 0.3 times the first frequency.

Disclosed is an automated walnut picking and collection method based on multi-sensor fusion technology, including: operation 1.1: when a guide vehicle for automated picking and collection is started, performing path planning for the guide vehicle; operation 1.2: remotely controlling the guide vehicle to move in a park according to a first predetermined rule, and collecting laser data of the entire park; operation 1.3: constructing a two-dimensional offline map; operation 1.4: marking a picking road point on the two-dimensional offline map; operation 2.1: performing system initialization; operation 2.2: obtaining a queue to be collected; operation 2.3: determining and sending, by the automated picking system, a picking task; operation 2.4: arriving, by the picking robot, at picking target points in sequence; operation 2.5: completing a walnut shaking and falling operation; and operation 2.6: collecting shaken walnuts. The provided method can obtain high-precision fruit coordinates and complete autonomous harvesting precisely and efficiently.

Systems and methods are disclosed herein for optimizing harvester yield. In an embodiment, a controller receives a pre-harvest image from a front-facing camera of a harvester. The controller inputs the pre-harvest image into a model, and receives as output from the model a predicted harvest yield. The controller receives, from an interior camera of the harvester, a post-harvest image including the plants as harvested. The controller inputs the post-harvest image into a second model and receives, as output, an actual harvest yield of the plants as-harvested. The controller determines that the predicted harvest yield does not match the actual harvest yield, and outputs a control signal.

An apparatus includes a processing device configured to obtain time series diagnostic data associated with assets in an information technology (IT). The processing device is also configured to generate first modality information comprising behavior labels assigned to each of a plurality of time periods, a given behavior label for a given time period being based at least in part on measured feature values for the features collectively in the given time period. The processing device is further configured to generate second modality information comprising feature deltas characterizing differences between measured feature values for interdependent feature pairs. The processing device is further configured to perform multi-modal analysis of the time series diagnostic data to detect behavior patterns in the utilizing a machine learning model trained using the first modality information and the second modality information, and to initiate remedial action in the IT infrastructure responsive to detecting an anomalous behavior pattern.

Systems and methods are described for determining an actual pose of an articulating boom arm using an artificial intelligence mechanism (e.g., a neural network) trained to determine the actual pose of the articulating boom arm based on captured image data. In some implementations, an electronic processor is configured to control movement of the articulating boom arm based at least in part on pose information determined by applying the image-based neural network. In some implementations, an electronic processor is configured to train the neural network by using, as training data, captured image data and output signals from sensors indicative of measured positions of the components of the articulating boom arm.

Making failure scenarios using adversarial reinforcement learning is performed by storing, in a first storage, a variety of first experiences of failures of a player agent due to an adversarial agent, and performing a simulation of an environment including the player agent and the adversarial agent. It also includes calculating a similarity of a second experience of a failure of the player agent in the simulation and each of the variety of first experiences in the first storage, and updating the first storage by adding the second experience as a new first experience of the variety of first experiences in response to the similarity being less than a threshold. Additionally, the use of adversarial reinforcement learning can include training the adversarial agent by using at least one of the plurality of first experiences in the first storage to generate an adversarial agent having diverse experiences.

The present disclosure pertains to autonomous control of an imaging system. In some embodiments, training information including at least a plurality of images and action information are received. The plurality of images and action information are provided to a prediction model to train the prediction model. Further, an image capturing device is controlled to capture an image of a portion of a living organism, the image is processed, via the prediction model, to determine an action to be taken with respect to the image, and the determined action is taken with respect to the image.