34638results about "Control devices" patented technology

The invention discloses an intelligent driving multi-sensor fusion data processing system, which belongs to the technical field of fusion data processing and comprises a data preprocessing unit, a data fusion unit and a data post-processing unit. The data preprocessing unit normalizes the original data of the laser radar, the camera, the millimeter wave radar and the ultrasonic sensor through a standardization, cleaning, noise reduction and calibration synchronization module; the feature level fusion module deeply digs the characteristics of each sensor, and fuses geometric, visual, motion and close-range obstacle features by using a deep neural network; the decision-making level fusion module generates a driving instruction through weighted voting and fuzzy logic reasoning through classification and situation evaluation; and the data post-processing unit combines the vehicle and road condition optimization instruction, evaluates the risk, and stores the data feedback optimization system. The system can improve the sensing precision and decision reliability, enhances the expansion adaptability of the system, and provides guarantee for safe and efficient operation of intelligent driving.

An example may receive at least one input via at least one device. An example may use the at least one input to determine an entity identity. An example may use the entity identity to create an automated agent and load context data associated with the entity identity into at least one layer of a multi-layer memory of the automated agent. An example may cause the automated agent to machine-learn a supervision level via the context data. The machine-learned supervision level may indicate a level of supervision of the automated agent by an entity associated with the entity identity. An example may configure the automated agent to execute a task on behalf of the entity and in accordance with the machine-learned supervision level.

The invention relates to the field of intelligent driving, in particular to a driver dangerous behavior intervention system and method based on a space-time diagram neural network, and the method comprises the steps: obtaining a physiological signal, a driving posture and control behavior data through a multi-modal data collection module; constructing a dynamic graph structure and evaluating a driver risk state by using a space-time diagram neural network risk perception engine; the intelligent intervention execution module adopts a tactile, visual or vehicle control intervention strategy according to the evaluation result; the closed-loop optimization module monitors an intervention effect and updates a risk model and a strategy; according to the system, millimeter wave radar and multi-channel visual analysis are fused, the recognition accuracy is improved to 98.6%, dangerous events are predicted 15-30 seconds in advance through physiological signals and micro-expression analysis, and sufficient response time is provided for a driver.

The method is suitable for the technical field of automatic driving, an automatic driving large model framework based on 3D space-time perception and human-like decision reasoning is provided, through deep integration of cross-modal dynamic semantic patch embedding and multi-scale chain reasoning, the three-dimensional space-time semantic reasoning ability of an automatic driving system in a complex scene is effectively improved, and the automatic driving efficiency is improved. The efficient cooperative processing of the full-view scene video stream and the global aerial view is realized; meanwhile, through a multi-level space-time semantic self-adaptive data extraction normal form and semantic action representation generation based on rule driving and semantic action interaction, and in combination with a cross-modal task-oriented multi-level training matrix architecture, the model shows excellent performance in the aspects of scene understanding, decision transparency and end-to-end driving task optimization. The accuracy of the system in dynamic behavior deduction is effectively improved, the decision interpretability is improved, and powerful technical support is provided for global path planning and human-like decision making of automatic driving.

An intravascular blood pump (100) comprises a tip (110), a first region (120) with at least one blood through-opening (121), a flow cannula (130), a second region (140) with at least one blood through-opening (141), a motor-operated pump mechanism (150) and a conducting cable (170). At least in the region of the flow cannula (130), at least one electrical conductor track is realized by a surface coating structure.

A blood pump assembly can include various components such as a housing and a sensor configured to detect one or more characteristics of the blood. In some embodiments, the sensor can be coupled to the housing and can include a sensor membrane configured to deflect in response to a change in a blood parameter (e.g., pressure). The blood pump assembly can include a shield that covers at least a portion of the sensor membrane so as to protect the sensor from damage when the blood pump assembly is inserted through an introducer and navigated through the patient's vasculature and / or when the blood pump assembly is inserted into the heart in a surgical procedure. One or more protective layers can be deposited over the sensor membrane to prevent the sensor membrane from being dissolved through interactions with the patient's blood.

The invention relates to the field of road traffic, in particular to a multi-feature fusion fatigue driving detection method and a fatigue driving detection system. The method comprises the following steps: extracting facial features from a face image of a driver; extracting vehicle features from the vehicle driving parameters of the vehicle driven by the driver; and fusing the facial features and the vehicle features to judge whether the driver is in fatigue driving. Extracting facial features by designing a CNN model; extracting basic convolution features; extracting local convolution features; extracting global convolution features; performing pooling operation; aggregating global features; and carrying out dimensionality reduction mapping. A self-encoder is designed to extract vehicle characteristics; a symmetric deep neural network structure is adopted, and high-dimensional time sequence data is compressed to a low-dimensional potential space through nonlinear mapping; through combination and matching of the CNN model and the auto-encoder, the technical defects of feature redundancy, noise interference, information loss and suboptimal decision existing in an existing multi-feature fusion fatigue driving detection system are thoroughly solved.

A control method for drift driving a vehicle supports a drift entry assistance function and a drift maintenance assistance function while maximizing driver's drift driving experience. The control method includes determining, by a controller, whether or not drift entry assistance control is required based on vehicle driving state information, when the vehicle enters a drift mode, and performing, by the controller, the drift entry assistance control using a torque vectoring motor so that the vehicle reaches a drift driving state when determining that the drift entry assistance control is required.

Apparatus and methods are described including placing, into a subject's body, a blood-pump tube, with an impeller disposed within the blood-pump tube. At least one blood-pressure-measurement element extends to at least an outer surface of the blood-pump tube, such that the distal end of the blood-pressure-measurement element is in direct fluid communication with a bloodstream of the subject outside the blood-pump tube at a location that is proximal to blood-inlet opening(s) defined by the blood-pump tube. Blood is pumped through the blood-pump tube, using the impeller. Pressure of the bloodstream of the subject outside the blood-pump tube is measured by measuring blood pressure at the distal end of the left-ventricular blood-pressure-measurement element. Other applications are also described.

The provided are a federated reinforcement learning (FRL) end-to-end autonomous driving control system and method, as well as vehicular equipment, based on complex network cognition. An FRL algorithm framework is provided, designated as FLDPPO, for dense urban traffic. This framework combines rule-based complex network cognition with end-to-end FRL through the design of a loss function. FLDPPO employs a dynamic driving guidance system to assist agents in learning rules, thereby enabling them to navigate complex urban driving environments and dense traffic scenarios. Moreover, the provided framework utilizes a multi-agent FRL architecture, whereby models are trained through parameter aggregation to safeguard vehicle-side privacy, accelerate network convergence, reduce communication consumption, and achieve a balance between sampling efficiency and high robustness of the model.

A system (100) for intelligent fault detection and energy management in fuel cell hybrid electric vehicles (FCHEVs), comprising: a data acquisition unit (102) comprising sensors (102a) configured to acquire real-time operational data from vehicle components; an energy management unit (104) comprising a fuel cell unit configured to generate electricity from hydrogen and a battery unit configured to store and supply electrical energy; a regenerative braking system (106) in conjunction with an energy management unit (104) configured to recover and store energy during vehicle deceleration; a centralized processing unit (108) configured to perform functions of the system (100) based on data acquired by the sensors (102a), wherein the centralized processing unit (108) comprises the following: a neural network processor (108a) configured to analyze real-time operational data for fault prediction; an energy optimization processor (108b) configured to dynamically manage power distribution; a maintenance prediction processor (108c) configured to identify potential component failures; a communication interface (110) configured to enable data exchange between vehicle and infrastructure (V2I) and between vehicles (V2V); a driver interface (112) configured to display system status and receive user input; and a control unit (114) configured to perform corrective actions based on neural network analysis and energy optimization calculations.

A sensor data evaluation device may include an artificial neural network, configured to receive and process first testing data, and output a first testing data output, wherein the first testing data represent sensor data of a first sensor detecting a sensor input; and one or more processors, configured to receive sensor deficiency information representing a deficiency of a second sensor; generate second testing data, wherein the second testing data are based on the sensor deficiency information; implement the artificial neural network to receive and process the second testing data; and output from the output layer a second testing data output; determine a difference between the first testing data output and the second testing data output; and if the difference is outside of one or more predetermined ranges or does not satisfy one or more predetermined criteria, send a signal representing a result of the determined difference.

The invention relates to a method for operating an implanted, ventricular assist system (2), comprising the following steps: a) determine a first impedance parameter at a first point in time by means of the assist system (2), b) determine a second impedance parameter at a second point in time by means of the assist system (2), c) at least determine a change of the impedance parameter using the first impedance parameter and the second impedance parameter, or compare at least the first or second impedance parameter to a threshold value.

A battery containment construct is provided that includes a unitary tray having a bottom and a set of walls including a first side wall, a second side wall, a first end wall, and a second end wall. The set of walls extending from the bottom of the tray and defining a cavity within the tray. A cover removably attached to the plurality of walls of said tray opposite the bottom of said tray. Energy absorption components attached to the bottom of the tray external to at least one of the set of walls in some versions to enhance impact resistance. The construct has high strength, is light weight and well suited for containing and protecting energy cells or batteries that adjusted to impact resistance, fire resistance, and fluid penetration prevention.

A system and method that manages charging of Electric Vehicles (EVs) at a depot equipped with EV chargers. For each EV arriving at the depot, data is collected from the EV including a current state of charge (SoC), an SoC goal, a route estimate, and a charging schedule. A depot controller determines from the collected data a charging priority for the EV relative to other EVs at the depot. When charging is required to meet the SoC goal of the EV, a charger is selected for charging the EV and a charging start time and charging duration is determined by an optimization algorithm that concurrently optimizes a minimal total cost of electricity for total power transferred to the EV during charging while maximizing a probability of the EV leaving the depot with its SoC goal satisfied.

The invention relates to the technical field of motor train unit battery pack evaluation, and provides a motor train unit battery health state evaluation and life prediction method and system, and the method comprises the steps: building a database through voltage fluctuation and temperature data in a floating charge state, and analyzing a correlation mechanism between target parameters, such as a battery capacity retention ratio and active lithium stock, and internal chemical substance loss; respectively extracting driving behavior characteristics, temperature parameters and electrochemical impedance spectroscopy by adopting second-level, minute-level and hour-level time granularity, and aligning multi-frequency data by utilizing a dynamic time warping algorithm; constructing a graph structure model taking the electrochemical parameters as nodes and cross-level association as edges, and extracting a battery aging characteristic topological relation through a graph neural network; and finally, establishing a prediction model fused with multi-dimensional features, and realizing accurate prediction of the residual life and the capacity recession inflection point of the battery. And the multi-scale characteristics are combined with an electrochemical mechanism, so that the accuracy of battery health state evaluation under a complex working condition is improved.

The invention provides a driving behavior analysis method and system based on multi-modal sensor fusion, and the method comprises the steps: firstly obtaining a multi-source sensor data set of a target vehicle containing at least two different sensor source driving environment perception data, and carrying out the time-space alignment processing of the multi-source sensor data set, and generating a multi-source fusion data set; then, through a feature extraction layer of a preset behavior analysis network, a driving behavior feature set containing vehicle operation, environment interaction and behavior continuity features is extracted from the multi-source fusion data set, an anomaly recognition layer is used for carrying out anomaly recognition on the driving behavior feature set, and an anomaly recognition result is generated; and finally, generating driving strategy optimization data according to an abnormal recognition result, and feeding back the driving strategy optimization data to a vehicle control system to adjust a driving strategy. According to the method, driving behaviors can be comprehensively analyzed, abnormity can be timely found, the driving strategy is optimized, and the driving safety and efficiency are improved.

The invention discloses a fatigue driving monitoring and early warning method based on multi-dimensional feature fusion, and relates to the field of intelligent traffic and safe driving, and the method comprises the steps: carrying out the feature extraction of a collected face image of a vehicle driver through a feature extraction model based on a convolutional neural network, and obtaining a face key point; then calculating the eye closing rate, the mouth opening rate, the head posture data and the facial muscle data to finally obtain multi-dimensional features, and fusing the multi-dimensional features by using a multi-layer perceptron to obtain fused facial features; and then, a fatigue state detection model obtained based on recurrent neural network training is utilized to detect the facial feature time sequence at a plurality of continuous sampling moments so as to detect whether a fatigue driving state occurs and trigger alarm feedback. According to the method, the fatigue state of the driver can be identified and judged more accurately through combination of multi-dimensional data fusion and a time sequence analysis algorithm, and the monitoring and early warning accuracy and reliability are improved. And the system can keep high robustness even in an illumination change or complex driving environment.

Systems and methods are described herein relating to an intracardiac blood pump system comprising a pump, an elongate multi-lumen catheter, a compartment connected to the catheter, a first lumen, a conduit and a filter disposed within the conduit. At least one electronic element is located within the compartment. The conduit extends through an interface between the compartment and the catheter. The conduit is configured to allow gas to penetrate from a first end of the conduit through the first opening of the compartment to a second end of the conduit. The second end of the conduit is in fluid communication with a central lumen of the elongate multi-lumen catheter. The filter is configured to prevent egress of liquid from the multi-lumen catheter into the compartment after allowing flow of gas from the first end of the conduit through the second end of the conduit and into the multi-lumen catheter.

A method for determining the speed of sound in a fluid in the region of an implanted, vascular support system includes sending an ultrasonic signal using an ultrasonic sensor, reflecting the ultrasonic signal on at least one sound reflector, receiving the reflected ultrasonic signal, and determining the speed of sound in the fluid using the ultrasonic signal. The at least one sound reflector may be visible in the field of vision of the ultrasonic sensor and arranged at a defined distance at least to the ultrasonic sensor or to a further sound reflector.

Blood pump assemblies and methods of manufacturing and operating blood pump assemblies are provided. The blood pump assembly includes a pump and an impeller blade rotatably coupled to the pump. The blood pump assembly also includes a pump housing component sized for passage through a body lumen and coupled to the pump. The pump housing component includes a peripheral wall extending about a rotation axis of the impeller blade. The peripheral wall includes an inner peripheral wall surface and an outer peripheral wall surface. The peripheral wall also includes one or more blood exhaust apertures. Each blood exhaust aperture in the one or more blood exhaust apertures is defined by an inner aperture edge and an outer aperture edge. Each inner aperture edge is chamfered between the inner peripheral wall surface and the outer peripheral wall surface.

Techniques for clustering sensor data are discussed herein. Sensors of a vehicle may detect data points in an environment. Clustering techniques can be used in a vehicle safety system to determine connection information between the data points. Further, the machine learned model can be trained to output motion data representing velocity data or a cluster association score. The connection information and / or motion data can be used by a vehicle computing device to detect objects in an environment and / or to control operation of a vehicle.

Examples relate to vehicle battery systems comprising a base unit connectable to a vehicle's electrical system, a removable modular replaceable battery, and a secondary battery within the base unit. The base unit is configured to interchangeably use modular replaceable batteries of different chemistries. Control circuits optimize charging and discharging parameters based on battery chemistries. The secondary battery provides power when the modular battery is removed. The system includes thermal management, compatibility with lead-acid charging systems, and mobile base unit functionality. The modular battery can power external devices and can foreseeably store energy up to and beyond 1 kWh. The system operates at various voltages suitable for different vehicle types.

The present disclosure relates generally to flight control of electric aircraft and other powered aerial vehicles. In one embodiment, a method is disclosed, comprising: receiving a descent rate command from a pilot input device, determining a proximity of each propeller of at least two propellers to a vortex ring state; and controlling the aircraft's descent rate to be less than the commanded descent rate when at least one of the at least two propellers is within a first threshold proximity to the vortex ring state.

One embodiment of a method for controlling a vehicle includes processing sensor data associated with a vehicle using a first trained machine learning model to generate one or more bird's eye view (BEV) features, performing one or more cross-attention operations based on the one or more BEV features and a learned embedding to generate an updated embedding, processing the updated embedding using a second trained machine learning model to generate one or more trajectories, and performing one or more operations to control the vehicle based on the one or more trajectories.

The invention relates to the technical field of new energy automobile thermal management, in particular to a direct cooling and heating management system and method in an automobile power battery. The method comprises the following steps: acquiring vehicle real-time operation condition data and battery pack state monitoring data; working condition mode recognition processing is conducted according to the vehicle real-time operation working condition data, and vehicle working condition mode feature data are generated; performing thermal load demand analysis on the battery pack state monitoring data based on the vehicle working condition mode characteristic data to obtain dynamic thermal load demand data of the battery pack; and according to the dynamic thermal load demand data of the battery pack, refrigerant demand quantity calculation and analysis are carried out, the refrigerant supply capacity of the current direct cooling system is evaluated, and refrigerant supply and demand matching feature data are obtained. The specific requirements of three heat sources of the battery, the motor and the cockpit for the refrigerant flow are quantified, and a refrigerant distribution strategy is formed according to the proportional relation, so that the safety of the battery is improved, and the driving comfort and the working stability of the electric driving system are also guaranteed.

Various systems, devices, and methods are disclosed herein for treating acute myocardial infarction (AMI) patients using a heart pump controlled in a manner that maximizes mechanical unloading of the left ventricle in the presence of cardiovascular instability and minimizes myocardial oxygen consumption (MVO2) and consequentially infarct size to prevent the development of subsequent heart failure. In a closed feedback system, the system can include a sensor configured to generate an output used to measure or calculate a left ventricular systolic pressure (LSVP) within the left ventricle of a heart and a controller coupled to a heart pump. The controller can be configured to measure or calculate the LVSP based on the output of the sensor and to control an operation of the heart pump to maximize mechanical unloading of the left ventricle based on the measured or calculated LVSP.

The invention relates to a method for determining at least a flow velocity or a fluid volume flow (5) of a fluid flowing through an implanted vascular support system (1), comprising the following steps: a) carrying out a pulsed Doppler measurement by means of an ultrasonic sensor (2) of the support system (1), b) evaluating a measurement result from step a), which has a possible ambiguity, c) providing at least one operating parameter of a flow machine (3) of the support system (1), d) determining at least the flow velocity or the fluid volume flow (5) using the measurement result evaluated in step b), wherein the possible ambiguity of the measurement result is corrected using the operating parameter.

The invention provides an electric bicycle charging method and system based on a mutual recognition mechanism, and relates to the technical field of battery management, and the method comprises the steps: detecting the state of a battery through a multi-sensor fusion algorithm, predicting the endurance mileage and charging demands through deep reinforcement learning, and predicting the destination and arrival time through track analysis; a charging pile is optimized based on a multi-constraint path planning model and an improved multi-target genetic algorithm, a target charging pile adopts dual encryption identity authentication, a self-adaptive charging control algorithm based on a fuzzy neural network is used for optimizing a charging process, and data is uploaded to a charging scheduling server.

A medical device with an implantable blood pump and a control and sensing unit configured to determine the flow rate generated by the blood pump when driven by an electric motor, wherein the flow rate is determined using peak-to-peak current data generated by the electric motor and, in some cases, associated heart rate data. In some embodiments, the validity of pulsatility of the resulting blood flow is determined and, if out of predetermined limits, an alarm may be actuated.