Sony Group Corporation

Sony Europe B.v.

Sony Corporation

Sony Network Communications Europe B.v.

Peter Exner

Hannes Bergkvist

Anders Isberg

Shaun Lee

Pär Spjuth

Ola Karl Thörn

Magnus Ola Johansson

Magnus Tillgren

Henrik Sundström

Andrej Petef

Mattias Falk

Thomas Fänge

Peter Blomqvist

A method for updating a neural network model on a computation device. The method includes estimating a bandwidth for data download from a server device to the computation device; estimating a time point of available computation capacity of the computation device; computing a maximum partition size as a function of the bandwidth and the time point; and causing download of a selected partition of the neural network model from the server device to the computation device, the selected partition being determined based on the maximum partition size. The method enables the selected partition to be executed by the computation device upon downloading and thereby provides for seamless updating of the neural network model while it is being executed on the computation device.

Updating a neural network model on a computation device

A remote control device is provided. The remote control device comprises a movement sensor, memory circuitry, a wireless interface, and processor circuitry configured to operate according to a movement model for identifying movement of the remote control device. The processor circuitry is configured to obtain, from the movement sensor, movement data indicative of movement of the remote control device. The processor circuitry is configured to determine, based on the movement data and by using the movement model, a user parameter indicative of a user of the remote control device. The processor circuitry is configured to determine whether the user parameter satisfies a first criterion. The processor circuitry is configured to, when the user parameter satisfies the first criterion, output the user parameter to a display device.

Remote control device, related devices, and related methods for identifying a user of a remote control device

An electronic device that includes a power source and a wireless transmitter is controlled by a method. The method predicts a magnitude of surplus energy in the power source under the assumption that the electronic device is operated in accordance with a default schedule over a time period. The default schedule defines sampling of sensor data from at least one sensor and transmission of the sensor data by use of the wireless transmitter. The method further determines, based on the magnitude of surplus energy, a modified schedule that results in sampling of an increased amount of sensor data per unit time compared to the default schedule, and configures the electronic device to operate in accordance with the modified schedule for at least part of the time period. The method increases availability of sensor data without compromising the mission of the electronic device.

Controlling an electronic device to sample and transmit sensor data

An electronic device, having a primary function different from detecting a fire, is configured for detecting a fire. The device includes processor circuitry, and a temperature sensor configured to generate temperature data, wherein the processor circuitry is configured to determine whether a first generated temperature data meets a first temperature criteria at a first time, and determine whether a second generated temperature data meets a second temperature criteria at a second time after the first time.

Electronic device and related methods for fire detection

A device is configured to assist an individual during extravehicular activity. The device obtains and operates on first data indicative of a detection space around the individual, and second data indicative of a pose of the head of the individual relative to a helmet, which is worn over the head of the individual with spacing to the head. The device is configured to process the first data to detect an object in the detection space and determine a first position of the object with respect to the helmet; determine, based on the first position and the second data, a second position of the object with respect to the head, and cause an array of speakers in the helmet to generate an audible sound inside the helmet with a spatial origin given by the second position.

Audio-based assistance during extravehicular activity

A computer system in included in a system for providing wildfire evacuation support. The computer system operates to obtain sensor data that represents a wildfire in a target region from a UAV which is deployed within the target region, obtain a current location of an individual in a hazardous situation in the target region, and obtain a desired destination for the individual in the target region. Based on the sensor data, the computer device identifies one or more danger zones that pose a fire-related threat to the individual, and determines at least one evacuation path that extends from the current location to the desired destination while avoiding the one or more danger zones. The computer system may be located at the UAV or on a central computer resource in communication with the UAV.

Computer system and method for providing wildfire evacuation support

An animal monitoring device includes memory circuitry, interface circuitry configured to communicate via short-range wireless communication to a second animal monitoring device and via cellular communication, and processor circuitry configured to obtain, from a first sensor, sensor data having a first transmission deadline and provide the sensor data into a sensor data set.

Animal monitoring device and related methods for monitoring an animal

A device is configured to provide assistance during extravehicular activity. The device obtains and operates on first data indicative of one or more objects in a detection space. The device is configured to process the first data for determination of one or more properties for the object(s); evaluate, based on the one or more properties, the object(s) in relation to a set of rules to determine one or more user instructions, and cause an illumination arrangement to provide the one or more user instructions by selective projection of the light in relation to the one or more objects. The device thereby reduces the reliance on cognitive processing by user(s) to take decisions on how to proceed in a situation, and instead the cognitive processing of information about the surroundings in relation to the task at hand is offloaded to the device.

Illumination-based assistance during extravehicular activity

A control system in a delivery system performs a method of controlling unmanned aerial vehicles, UAVs, which are organized in a group or swarm to perform one or more missions to deliver and/or pick up goods. The method includes obtaining drag data indicative of air resistance for one or more UAVs in the group, determining, as a function of the drag data, a respective relative position of at least the one or more UAVs within the group, and controlling at least the one or more UAVs to attain the respective relative position. Such adjustment in the relative position of one or more UAVs enables improved energy efficiency and may be made to reduce energy consumption, increase reach or decrease travel time of at least one UAV in the group.

Controlling a group of unmanned aerial vehicles for delivery of goods

A device is configured to provide performance support in a low-gravity environment. The device obtains and operates on first data indicative of a body pose of an individual, and second data indicative of a gaze direction of the individual. The device is configured to determine, based on the first data and the second data, a first time series of body poses and a second time series of gaze directions that represent the individual performing a task, obtain a nominal performance scheme for the task, perform an evaluation of the first and second time series in relation to the nominal performance scheme for detection of a performance deviation, and provide, based on the evaluation, feedback data for presentation by a feedback device.

Performance-based feedback for activity in a low-gravity environment

A scheduling device performs a method of scheduling transports by a fleet of unmanned aerial vehicles, UAVs, to enable improved utilization of the fleet. The method includes computing, for the fleet, aggregated cost data, ACM, that associates utilization cost with distributed sub-regions within a geographic region, the utilization cost of a respective sub-region representing an estimated cost for directing at least one of the UAVs to the respective sub-region. The method further includes receiving a query indicative of one or more potential locations for pick-up or delivery of a payload, determining, based on the ACM, a transportation price for at least one potential location, and presenting the transportation price for the at least one potential location. The scheduling device thereby provides an interactive ordering system that enables utilization of the fleet to be automatically optimized by swarm intelligence.

Utilization of a fleet of unmanned aerial vehicles for delivery of goods

A monitoring system detects and mitigates traffic risks among a group of vehicles. The group of vehicles includes a ground-based vehicle (GBV), e.g. an automotive vehicle, and an air-based vehicle (ABV), e.g. a drone, which is operated to track a ground-based object (GBO), e.g. an unprotected road user or an animal. The monitoring system performs a method comprising: obtaining () predicted navigation data for the ground-based vehicle and the air-based vehicle, processing () the predicted navigation data to obtain one or more future locations of the ground based-object and to detect an upcoming spatial proximity between the ground-based object and the ground-based vehicle, and causing (), upon detection of the upcoming spatial proximity, an alert signal to be provided to at least one of the ground-based object and the ground-based vehicle.

Monitoring vehicle movement for traffic risk mitigation

A control system in a delivery system performs a method to deliver a payload by an unmanned aerial vehicle, UAV, without requiring presence of the recipient even if the payload comprises sensitive or vulnerable goods. The method comprises obtaining a designated location for delivery of the payload, obtaining at least one payload vulnerability index for the payload, and obtaining a pick-up time indicative of a time point when the payload is expected to be retrieved by the recipient. The method further comprises selecting a final delivery area at or near the designated location as a function of the at least one payload vulnerability index and the pick-up time, and operating the UAV to deliver the payload to the final delivery area.

Safeguarded delivery of goods by unmanned aerial vehicles

A portable electronic device is controlled, for example to selectively activate a flight mode, based on first event data and/or second event data. The first event data is generated by operating a first model for detection of airplane flight events, AFEs, on first sensing data representing ambient pressure. The second event data is generated by operating a second model for detection of AFEs on second sensing data representing own motion or ambient sound. A control method in the portable electronic device generates training data comprising groups of time-aligned data samples from the first event data and the second sensing data, generates an updated second model by use of the training data, and replaces the second model by the updated second model. The control method facilitates adaptation of the portable electronic device to detect airplane flight events in new environments and enables improved robustness when detecting AFEs.

Operating a portable electronic device to detect airplane flight events

A first electronic device includes an interface; microphone circuitry; memory circuitry; and processor circuitry. The first electronic device is configured to discover, via the interface, a first wireless network. The first electronic device is configured to receive, via the interface, from a second electronic device discovering the first wireless network, a voice biometric indicative of a second user. The first electronic device is configured to activate the microphone circuitry to detect an input audio signal. The first electronic device is configured to determine whether the detected input audio signal satisfies one or more criteria. The first electronic device is configured to, when the detected input audio signal satisfies the one or more criteria, determine a parameter indicative of a level of risk of exposure; and generate, based on the parameter, contact data.

Growing community of inventors

Malmö, Sweden

Peter Exner