Abstract: Imitation and reinforcement learning for multi-agent simulation includes performing operations. The operations include obtaining a first real-world scenario of agents moving according to first trajectories and simulating the first real-world scenario in a virtual world to generate first simulated states. The simulating includes processing, by an agent model, the first simulated states for the agents to obtain second trajectories. For each of at least a subset of the agents, a difference between a first corresponding trajectory of the agent and a second corresponding trajectory of the agent is calculated and determining an imitation loss is determined based on the difference. The operations further include evaluating the second trajectories according to a reward function to generate a reinforcement learning loss, calculating a total loss as a combination of the imitation loss and the reinforcement learning loss, and updating the agent model using the total loss.

Abstract: Diffusion for realistic scene generation includes obtaining a current set of agent state vectors and a map data of a geographic region, and iteratively, through multiple diffusion timesteps, updating the current set of agent state vectors. Iteratively updating includes processing, by a noise prediction model, the current set of agent state vectors, a current diffusion timestep of the plurality of diffusion timesteps, and the map data to obtain a noise prediction value, generating a mean using the noise prediction value, generating a distribution function according to the mean, sampling a revised set of agent state vectors from the distribution function, and replacing the current set of agent state vectors with the revised set of agent state vectors. The current set of agent state vectors are outputted.

Abstract: An automatic moving snow removal device including a moving module, driving a snow blower to move; a working module, including a working motor and a snow throwing mechanism driven by the working motor, the snow throwing mechanism is driven by the working motor to collect accumulated snow and inclusions on the ground and throw out of the snow throwing mechanism; and a control module, configured to control a rotary speed of the working motor to cause a speed when the inclusions depart from the snow throwing mechanism is not higher than 41 m/s.

Abstract: A returning method of a self-moving device, a self-moving device are provided. In the returning method, a self-moving device autonomously moves inside a working region based on a map. Specifically, the method includes: acquiring a current position of the self-moving device in the working region; selecting a return path to a target position according to the current position; determining a reuse status of the return path, determining, based on the reuse status of the return path, whether to reselect a return path; and enabling the self-moving device to return to the target position along the selected return path.

Abstract: A device and method used to image conduits, such as pipes, wellbores and tubulars, with imaging sensors, such as cameras and ultrasound arrays. The speed and location of the device are determined using one or more speed sensor modules. Images are then registered to more precise axial locations along the conduit than are normally possible using wireline encoders or other methods. The conduit may be visualized to proper scale for improved analysis of defects.

Abstract: An automatic moving snow removal device including a moving module, driving a snow blower to move; a working module, including a working motor and a snow throwing mechanism driven by the working motor, the snow throwing mechanism is driven by the working motor to collect accumulated snow and inclusions on the ground and throw out of the snow throwing mechanism; and a control module, configured to control a rotary speed of the working motor to cause a speed when the inclusions depart from the snow throwing mechanism is not higher than 41 m/s.

Abstract: A method includes obtaining, from sensor data, map data of a geographic region and multiple trajectories of multiple agents located in the geographic region. The agents and the map data have a corresponding physical location in the geographic region. The method further includes determining, for an agent, an agent route from a trajectory that corresponds to the agent, generating, by an encoder model, an interaction encoding that encodes the trajectories and the map data, and generating, from the interaction encoding, an agent attribute encoding of the agent and the agent route. The method further includes processing the agent attribute encoding to generate positional information for the agent, and updating the trajectory of the agent using the positional information to obtain an updated trajectory.

Abstract: The present disclosure relates to a self-moving device. The self-moving device includes a movement mechanism disposed on a body of the self-moving device and configured to drive the self-moving device to move; a connecting arm connected to the body and at least one working head of a plurality of working heads configured to perform different work tasks; a safety detection module configured to detect whether at least one of the self-moving device or the at least one working head is in a safe working environment; and a controller configured to control a working status of at least one of the self-moving device or the at least one working head based at least in part on a detection result of the safety detection module.

Abstract: A charging system for executing a charging process during a charging frame is provided. The charging system includes one or more electric vehicles (EV) chargers; a load management system (LMS) capable of controlling the one or more EV chargers; a network system comprising one or more wired or wireless connections for linking the one or more EV chargers to the load management system; and a precision-time energy-based control algorithm for dynamically regulating an output power of the one or more EV chargers to meet a preset charging profile. The charging frame is divided into a plurality of sections and the LMS adjusts the output power of each section for compensating an energy shortage in previous sections.

Abstract: Disclosed is a novel CT denoising method using deep learning to boost the image signal-to-noise ratio for improved disease detection in patients with acute ischemic stroke (AIS). This method reduced the original CT image noise significantly better than BM3D, with SNR improvements in GM, WM, and DG by 2.47×, 2.83×, and 2.64× respectively and CNR improvements in DG/WM and GM/WM by 2.30× and 2.16× respectively. Scans denoised by the proposed model are shown to be visually clearer with preserved anatomy. The disclosed deep learning model significantly reduces image noise and improves signal-to-noise and contrast-to-noise ratios in 380 unseen head NCCT cases.

Abstract: A system includes a mobile device for capturing raw video of a subject, a preprocessing system communicatively coupled to the mobile device for splitting the raw video into an image stream and an audio stream, an image processing system communicatively coupled to the preprocessing system for processing the image stream into a spatiotemporal facial frame sequence proposal, an audio processing system for processing the audio stream into a preprocessed audio component, one or more machine learning devices that analyze the facial frame sequence proposal and the preprocessed audio component according to a trained model to determine whether the subject is exhibiting signs of a neurological condition, and a user device for receiving data corresponding to a confirmed indication of neurological condition from the one or more machine learning devices and providing the confirmed indication of neurological condition to the subject and/or a clinician via a user interface.

Abstract: A self-moving robot includes a shell, a driving module, driving the self-moving robot to move on the ground; a mowing module, executing mowing work; an energy module, providing energy for the self-moving robot; a control module, controlling the self-moving robot to automatically move and execute work, the self-moving robot further includes a cleaning module executing ground cleaning work; the self-moving robot has a mowing mode and a cleaning mode, under the mowing mode, the control module controls the self-moving robot to execute mowing work, and under the cleaning mode, the control module controls the self-moving robot to execute cleaning work.

Abstract: The present invention relates to a self-moving device (100). The self-moving device (100) includes a body (20), a walking mechanism (40) disposed on the body (20) and configured to drive the self-moving device (100) to walk, a connecting arm (60) connected to the body (20), and a control module (11) configured to control the walking mechanism (40) to drive the self-moving device (100) to walk within a defined area. The connecting arm (60) is selectively connected to at least one of at least two working heads (200) configured to perform different work tasks. The connecting arm (60) includes a connecting structure (64) configured to be connected to the working head (200) and a connecting component (62) configured to connect the connecting structure (64) to the body (20).

Abstract: A link interface is provided of a communication protocol using idle flow control digits (flits) to maintain link continuity. The link interface includes: a physical layer of the communication protocol configured to transmit and receive flits via a link, wherein the communication protocol provides for idle flits of first and second sizes for maintaining link continuity, the first size being smaller than the second size; and a data link layer configured to transmit and receive flits to/from the physical layer. The data link layer is configured to remove idle flits of the first size received from the physical layer and to report cyclic redundancy check errors of filtered first sized idle flits in a correct order in relation to other flits.

Abstract: An automatic moving snow removal device including a moving module, driving a snow blower to move; a working module, including a working motor and a snow throwing mechanism driven by the working motor, the snow throwing mechanism is driven by the working motor to collect accumulated snow and inclusions on the ground and throw out of the snow throwing mechanism; and a control module, configured to control a rotary speed of the working motor to cause a speed when the inclusions depart from the snow throwing mechanism is not higher than 41 m/s.

Abstract: Parity data associated with commands to, and indications from, a configuration register that includes a first command FIFO for receiving commands and a response FIFO for returning indications can be managed. Commands can be tracked by storing the commands in a second command FIFO and a command can be dequeued from the second command FIFO, in response to a command emerging from the response FIFO. Parity data can be generated from the data associated with a write operation, and stored in a parity latch corresponding to the configuration register, in response to the dequeued command being a successfully completed write operation. The generated parity data can be read from a parity latch corresponding to the configuration register and provided the generated parity data for return with an indication that the dequeued command is a successfully completed write operation.

Abstract: A device and method used to image conduits, such as pipes, wellbores and tubulars, with imaging sensors, such as cameras and ultrasound arrays. The speed and location of the device are determined using one or more speed sensor modules. Images are then registered to more precise axial locations along the conduit than are normally possible using wireline encoders or other methods. The conduit may be visualized to proper scale for improved analysis of defects.

Abstract: An automatic moving snow removal device including a moving module, driving a snow blower to move; a working module, including a working motor and a snow throwing mechanism driven by the working motor, the snow throwing mechanism is driven by the working motor to collect accumulated snow and inclusions on the ground and throw out of the snow throwing mechanism; and a control module, configured to control a rotary speed of the working motor to cause a speed when the inclusions depart from the snow throwing mechanism is not higher than 41 m/s.

Abstract: The embodiments relates to garden blower, including: an enclosure, including a main body part located on back end and a blowing pipe located on front end and extending axially, the enclosure disposed with an air inlet and an air outlet communicated with external environment; a power device, connected to the enclosure; a fan component, driven by the power device and generating airflows; the fan component includes at least two-level fans, further includes a first-level fan and a second-level fan axially disposed front and back, the garden blower includes a first air inlet passage allowing entrance of the airflows generated by the first-level fan and a second air inlet passage allowing entrance of the airflows generated by the second-level fan, and the airflows entering the first air inlet passage and the airflows entering the second air inlet passage are converged into the blowing pipe and are blown to the outside.

Abstract: A returning method of a self-moving device, a self-moving device are provided. In the returning method, a self-moving device autonomously moves inside a working region based on a map. Specifically, the method includes: acquiring a current position of the self-moving device in the working region; selecting a return path to a target position according to the current position; determining a reuse status of the return path, determining, based on the reuse status of the return path, whether to reselect a return path; and enabling the self-moving device to return to the target position along the selected return path.