Abstract: Aspects of the present disclosure provide an environmental probe configured to be at least partially inserted into the ground at a location and to provide defensible space monitoring and maintenance for a home or other structure. The environmental probe may include one or more environmental sensors configured to perform various environmental measurements to generate environmental measurement data. The environmental probe (e.g., a processor of the environmental probe) may compare the environmental measurement data to one or more thresholds to determine an alert state associated with the location. The environmental probe may include one or more wireless interfaces configured to enable communication with a remote device, such as a smart hub device or other environmental probes. The environmental probe may transmit an indicator of the alert state to the remote device to enable performance of one or more operations based on the alert state.

Abstract: This document describes self-cleaning devices. In one aspect, a self-cleaning device includes a fabric having a surface covered with a photocatalyst, one or more light sources embedded in the fabric, and a triggering mechanism that activates a cleaning cycle by activating the one or more light sources. The triggering mechanism can include a pressure sensor. The triggering mechanism can be configured to activate the cleaning cycle in response to detecting a decrease in pressure being applied to the pressure sensor.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for recognizing speech using a spiking neural network acoustic model implemented on a neuromorphic processor are described. In one aspect, a method includes receiving, a trained acoustic model implemented as a spiking neural network (SNN) on a neuromorphic processor of a client device, a set of feature coefficients that represent acoustic energy of input audio received from a microphone communicably coupled to the client device. The acoustic model is trained to predict speech sounds based on input feature coefficients. The acoustic model generates output data indicating predicted speech sounds corresponding to the set of feature coefficients that represent the input audio received from the microphone. The neuromorphic processor updates one or more parameters of the acoustic model using one or more learning rules and the predicted speech sounds of the output data.

Abstract: This document describes neuromorphic controllers. In one aspect, a method for controlling one or more joints of a robotic arm includes receiving, by neuromorphic controller comprising a spiking neural network (SNN), a target value of a joint control variable for a joint of the robotic arm. The SNN includes two position proprioceptor neurons, two-speed proprioceptor neurons, a presynaptic inhibitory neuron, an extensor motor neuron, and a flexor motor neuron. The neuromorphic controller updates an actual value of the joint control variable for the joint of the robotic arm based on the target value of the joint control variable. The updating includes generating, by one of the two position proprioceptor neurons, the first spikes to one of the extensors motor neurons or the flexor motor neuron based on a difference between the actual value of the joint control variable and the target value of the joint control variable.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for updating a trained gesture recognition model deployed on a neuromorphic processor that has been trained to process data that characterizes the new gesture and to determine a gesture classification for the gesture are described. A method includes receiving data that characterizes a new gesture and processing the data to generate a new embedding in a latent space. For each of multiple clusters of reference embeddings in the latent space, a respective distance in the latent space between the cluster of reference embedding and the new embedding is determined. A determination is made, based on applying one or more learning rules to the distances, one or more procedures to update the gesture recognition model. A determination is made, in accordance with the determined procedure(s), an update to values of one or more parameters of the gesture recognition model.

Abstract: Electrical input devices can be produced using a multi-material 3D-printing process. The electrical input devices can include a non-conductive material portion and a conductive material portion. The non-conductive and conductive material portions are integrally formed during a single 3D-printing process. Deformation of the electrical input devices cause an electrical variance of the conductive material portion that is responsive to the deformation. Some electrical input devices described provide digital responses, and some electrical input devices described provide analog responses. The described techniques can be used to manufacture complex finished devices in a single 3D-print run, and, in some examples, without the need for post-processing or assembly.

Abstract: An upper-extremity prosthetic is adapted to engage with an athletic ball. The prosthetic includes one or more springs that provide energy return as a user is throwing the ball using the prosthetic. The springs can have a conductivity that changes in relation to an amount of strain or deformation of the spring. The change in conductivity can be used to provide haptic feedback to the user so the user can sense the amount of force being applied to throw the ball. In some embodiments, the springs are made by a multi-material 3D printing (additive manufacturing) process and include a first material that is electrically non-conductive and a second material that electrically conductive. In some embodiments, the prosthetic also includes one or more cantilevered springs that are also adapted to engage with the ball and to provide energy return while throwing the ball.

Abstract: Springs can provide energy return and have a conductivity that changes in relation to an amount of strain or deformation of the spring. An upper-extremity prosthetic device includes a first coil spring coupled to a first member and a first cantilever spring extending from the first member to a surface adapted to engage with an object. The first coil spring is arranged to absorb energy and to provide energy return in response to movement of the first member. The first coil spring includes a first conductive surface and a second conductive surface separate from the first conductive surface by non-conductive surfaces. The first cantilever spring includes a conductive trace with a plurality of conductive segments arranged on the conductive trace.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for recognizing speech using a spiking neural network acoustic model implemented on a neuromorphic processor are described. In one aspect, a method includes receiving, a trained acoustic model implemented as a spiking neural network (SNN) on a neuromorphic processor of a client device, a set of feature coefficients that represent acoustic energy of input audio received from a microphone communicably coupled to the client device. The acoustic model is trained to predict speech sounds based on input feature coefficients. The acoustic model generates output data indicating predicted speech sounds corresponding to the set of feature coefficients that represent the input audio received from the microphone. The neuromorphic processor updates one or more parameters of the acoustic model using one or more learning rules and the predicted speech sounds of the output data.

Abstract: The present disclosure relates to electronic devices and methods of manufacturing electronic devices. A method of manufacturing a dissolvable electronic device includes forming a dissolvable sheet; applying a self-sintering agent to the dissolvable sheet to form a substrate; and depositing electrically conductive ink onto the substrate in a trace. A method of manufacturing a meltable electronic device includes mixing a conductive material with a melted wax to form a conductive wax mixture in liquid form; molding the conductive wax mixture; and solidifying the conductive wax mixture to obtain the meltable electronic device. A method of manufacturing an edible electronic device includes cutting a layer of conductive material to form a pattern that defines a circuit; applying the layer of conductive material to an edible medium, wherein the edible medium is in liquid or semi-solid form; and solidifying the edible medium to obtain the edible electronic device.

Abstract: A method of producing a textile sensor includes: obtaining an organic fabric; carbonizing the organic fabric by applying heat to the organic fabric in an inert environment to form a conductive fabric; and attaching one or more electrical terminals to the conductive fabric. The method includes coating the conductive fabric with a polymeric encapsulating material. The method includes, for each of the one or more electrical terminals, connecting a first end of a flexible conductor to the electrical terminal and connecting a second end of each flexible conductor to a wireless interface printed circuit board. The textile sensor comprises at least one of a pressure sensor, a proximity sensor, a touch sensor, a strain sensor, a wind sensor, a temperature sensor, a heating element, a triboelectric sensor, and an energy harvester.

Abstract: This document described pressure-sensitive human machine interface (HMI) devices. In some aspects, a pressure-sensitive human-machine interface (HMI) device may include a surface element comprising an outer surface for receiving pressure inputs. The HMI device may also include multiple pressure-sensitive input regions that each include one or more positive electrodes and one or more return electrodes. The HMI device may also include a pressure-sensitive conductive sheet disposed adjacent to or between the one or more positive electrodes and one or more return electrodes of the multiple pressure-sensitive input regions.

Abstract: Springs can provide energy return and have a conductivity that changes in relation to an amount of strain or deformation of the spring. In some embodiments, the springs are made by multi-material 3D printing (additive manufacturing). Such springs made by multi-material 3D printing may include a first material that is electrically non-conductive and a second material that electrically conductive. The extent of deformation or strain of the spring may be determined or estimated by measuring the conductivity or resistivity of the electrically conductive material portion of the spring.

Abstract: Implementations are directed to a pressure-sensing device including a pressure-sensitive sheet, one or more pressure-sensitive input regions disposed along the pressure-sensitive sheet including a first conductive thread including a first length in contact with the pressure-sensitive sheet, and a second conductive thread including a second length in contact with the pressure-sensitive sheet. At least a first portion of the first length of the first conductive thread passes through the pressure-sensitive sheet through a first hole in the pressure-sensitive sheet at a first location and a second portion of the second length of the second conductive thread passes through the pressure-sensitive sheet through a second hole in the pressure-sensitive sheet at a second location.

Abstract: This document describes morphing objects that are created using hydrogels and techniques for creating and controlling the morphing objects. In one aspect, a method for creating a morphing object includes creating one or more heating elements on a substrate by applying one or more conductive traces onto at least one of a first surface of the substrate or a second surface of the substrate opposite the first surface and forming a hydrogel pattern on the first surface of the substrate by applying a hydrogel ink to the first surface of the substrate based on a predefined pattern.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for converting audio to spikes for input to a spiking neural network configured to recognize speech based on the spikes are described. In some aspects, a method includes obtaining audio data and generating frequency domain audio signals that represent the audio data by converting the audio data into a frequency domain. The frequency domain audio signals are mapped into a set of Mel-frequency bands to obtain Mel-scale frequency audio signals. A log transformation is performed on the Mel-scale frequency audio signals to obtain log-Mel signals. Spike input is generated for input to a spiking neural network (SNN) model by converting the log-Mel signals to the series of spikes. The spike input is provided as an input to the SNN model.

Abstract: Implementations are directed to a smart filter apparatus including an air filter, a sensor supported by the air filter and sensitive to a pathogen, and a control unit in data communication with the sensor and operable to perform operations including receiving, from the sensor, an indication of the pathogen present in air incident on the sensor, generating, by the control unit, a pathogen alert response, and providing, by the control unit, the pathogen alert response.

Abstract: Electrical input devices, conductive traces, and microcontroller interface devices can be created in a single print using a multi-material 3D printing process. The devices can include a non-conductive material portion and a conductive material portion. The non-conductive and conductive material portions are integrally formed during a single 3D printing process. For example, a fully functional QWERTY keyboard, ready to receive a microcontroller, can be multi-material 3D printed using the techniques described herein.

Abstract: Aspects of the present disclosure provide an environmental probe configured to be at least partially inserted into the ground at a location and to provide defensible space monitoring and maintenance for a home or other structure. The environmental probe may include one or more environmental sensors configured to perform various environmental measurements to generate environmental measurement data. The environmental probe (e.g., a processor of the environmental probe) may compare the environmental measurement data to one or more thresholds to determine an alert state associated with the location. The environmental probe may include one or more wireless interfaces configured to enable communication with a remote device, such as a smart hub device or other environmental probes. The environmental probe may transmit an indicator of the alert state to the remote device to enable performance of one or more operations based on the alert state.

Abstract: Aspects of the present disclosure provide an environmental probe configured to be at least partially inserted into the ground at a location and to provide defensible space monitoring and maintenance for a home or other structure. The environmental probe may include one or more environmental sensors configured to perform various environmental measurements to generate environmental measurement data. The environmental probe (e.g., a processor of the environmental probe) may compare the environmental measurement data to one or more thresholds to determine an alert state associated with the location. The environmental probe may include one or more wireless interfaces configured to enable communication with a remote device, such as a smart hub device or other environmental probes. The environmental probe may transmit an indicator of the alert state to the remote device to enable performance of one or more operations based on the alert state.