An underwater work system of the present disclosure acquires a relative position of an underwater vehicle relative to a surface ship at the start of searching work, the relative position being measured based on a sound wave transmitted from a wave transmitter. The underwater work system calculates a position of the underwater vehicle based on the acquired relative position. When a measurement error region whose center corresponds to the calculated position of the underwater vehicle and an expected laid region of a pipeline extending in a predetermined direction overlap each other, the underwater work system moves the underwater vehicle to such a position that the measurement error region and the expected laid region do not overlap each other, and then, makes the underwater vehicle perform crossing detection in which the underwater vehicle detects the presence or absence of the pipeline while crossing the expected laid region.

Conventional gesture detection approaches demand large memory and computation power to run efficiently, thus limiting their use in power and memory constrained edge devices. Present application/disclosure provides a Spiking Neural Network based system which is a robust low power edge compatible ultrasound-based gesture detection system. The system uses a plurality of speakers and microphones that mimics a Multi Input Multi Output (MIMO) setup thus providing requisite diversity to effectively address fading. The system also makes use of distinctive Channel Impulse Response (CIR) estimated by imposing sparsity prior for robust gesture detection. A multi-layer Convolutional Neural Network (CNN) has been trained on these distinctive CIR images and the trained CNN model is converted into an equivalent Spiking Neural Network (SNN) via an ANN (Artificial Neural Network)-to-SNN conversion mechanism. The SNN is further configured to detect/classify gestures performed by user(s).

The invention relates to a method for evaluating multiple received signals (6, 8, 11), wherein the method has the following steps: sending a transmission signal (3) receiving a first signal (6), which contains the transmission signal via a first receiver (5) and receiving a second signal (8), which contains the transmission signal via a second receiver (7), characterized in that to evaluate the received signals (6, 8), the received signals (6, 8) are compared with one another, wherein the comparison comprises determination of a time difference and/or phase difference between the first signal (6) and the second signal (8).

System, computer products, and methods can improve the resolution of data from a sensor array. One of these methods include receiving, from an analog to digital converter, a series of measurements representing frequency samples and spatial samples from a sensor array. The method includes generating a plurality of factors based on a polynomial. The method includes applying one or more complex weights to the measurements based on the factors. The method includes combining the complex weighted measurements into a plurality of values. The method also includes identifying a characteristic of an object detected by the sensor array based on the plurality of values.

An in-vehicle object determining apparatus cooperates with an obstacle sensor unit, which detects an obstacle at a first time. An estimated detected state is calculated as a detected state of the obstacle estimated to be detected by the obstacle sensor unit at a second time after a lapse of a predetermined time period from the first time, on condition that the obstacle is assumed to be under stationary state, based on (i) a vehicle-relative position of the obstacle detected at the first time, (ii) a sensor position of the obstacle sensor unit, and (iii) a vehicle position change during a period from the first time to the second time. It is determined that the obstacle is a moving object based on a discrepancy between the estimated detected state of the obstacle and a real detected state of the obstacle actually detected by the obstacle sensor unit at the second time.

The present invention relates to a proximity detection system and method for a computer or similar device. The computer includes at least two transducer units being capable of transmitting acoustic signals identifying the transducer within a predetermined frequency range. The transducer units are positioned at a distance from each other in said device, and at least one receiving transducer is capable of receiving acoustic signals within said frequency range and recognizing the transducer identification. The system is configured to measure the distance between each transducer unit and a reflecting object based on the transmitted and received signals based on the measured propagation time. The system is further configured to activate each of the transducer units and comparing the measured distance from each transducer unit thus calculating the direction of a reflecting object, and at the detecting of an object within a predetermined range of direction providing an activation signal to the device.

An object sensing apparatus includes an object sensing unit to sense an object using a direct wave, being a reflected wave received by a sensor having transmitted a probing wave among a plurality of ultrasonic sensors, and an indirect wave, being a reflected wave received by a sensor different from the sensor having transmitted the probing wave among the ultrasonic sensors, and a temperature change detection unit to detect a predetermined temperature change state where temperature change of a predetermined value or more has occurred or a possibility of the temperature change occurs in an ambient temperature of the moving object. When the predetermined temperature change state is detected by the temperature change detection unit, the object sensing unit performs sensing suppression control not to sense the object using the indirect wave or to make it difficult to sense the object using the indirect wave.

A plurality of control points (10) are defined which each have a known spatial relationship relative to an array of transducers. An amplitude is assigned to each control point (12). A matrix (16) is produced containing elements which represent, for each of the control points, the effect that producing a modeled acoustic field having the assigned amplitude with a particular phase at the control point has on the consequential amplitude and phase of the modeled acoustic field, at the other control points (14). Eigenvectors of the matrix (18) are determined, each eigenvector representing a set of phases and relative amplitudes of the modeled acoustic field at the control points. One of the sets (20) is selected and the transducer array is operated to cause one or more of the transducers to output an acoustic wave each having an initial amplitude and phase such that the phases and amplitudes of the resultant acoustic field at the control points correspond to the phases and relative amplitudes of the selected set (22, 24).

A method for non-coherent stereo radar tracking includes, at a stereo radar system, transmitting a probe signal, receiving a reflected probe signal in response to reflection of the probe signal by a tracking target, calculating first and second target ranges from the reflected probe signal data, transforming the reflected probe signal data based on the first and second target ranges, and calculating a first target angle from the transformed reflected probe signal data.

Automotive radar systems may employ a reconfigurable connection of antennas to radar transmitters and/or receivers. An illustrative embodiment of an automotive radar system includes: a radar transmitter; a radar receiver; and a digital signal processor coupled to the radar receiver to detect reflections of a signal transmitted by the radar transmitter and to derive signal measurements therefrom. At least one of the radar transmitter and the radar receiver are switchable to provide the digital signal processor with signals from each of multiple combinations of transmit antenna and receive antenna.