Techniques and apparatuses are described that implement motion classification using low-level detections. In particular, a radar system identifies fused detections associated with an object and determines whether the fused detections indicate that the object is moving. If it is determined to be moving or moving perpendicular to the host vehicle, a current motion counter or perpendicular motion counter is incremented, respectively. A current motion flag and/or a perpendicular motion flag are set as true if the current motion counter or the perpendicular motion counter has a value greater than a threshold value, respectively. In response to setting either flag as true, the radar system increments a historical motion counter as true. The host vehicle is then operated based on the current motion flag, the perpendicular motion flag, and the historical motion counter. In this way, the radar system introduces hysteresis to improve the reliability and stability of motion classification.

Systems and methods for Wi-Fi sensing using UL-OFDMA are provided. Wi-Fi sensing systems include sensing devices and sensing transmitters configured to communicate through radio-frequency signals. Initially, first channel resources are allocated to first expected transmissions from the sensing transmitters and first sensing trigger message to trigger first series of sensing transmissions from the sensing transmitters is transmitted. Further, a first series of sensing transmissions is received, and the first series of sensing measurements are generated. Thereafter, identification of feature of interest is obtained and a selection of sensing transmitters is determined. Second channel resources are allocated to second expected transmissions from the selection of sensing transmitters. A second sensing trigger message to trigger a second series of sensing transmissions from the selection of the sensing transmitters is provided. A series of sensing transmissions is received, and a second series of sensing measurements is generated based on the second series of sensing transmissions.

A system can include a computer including a processor and a memory, the memory storing instructions executable by the processor to receive point cloud data. The instructions further include instructions to generate a plurality of feature maps based on the point cloud data, each feature map of the plurality of feature maps corresponding to a parameter of the point cloud data. The instructions further include instructions to aggregate the plurality of feature maps into an aggregated feature map. The instructions further include instructions to generate, via a feedforward neural network, at least one of a segmentation output or a classification output based on the aggregated feature map.

A handling device (103) for loads (105) includes a radar transmitter (107), a radar sensor (111) and a data-processing device. The radar transmitter (107) is configured to irradiate at least part of the handling device (103) and/or the load (105). The radar sensor (111) is configured to detect at least part of the irradiated handling device (103) and/or the irradiated load (105). The data-processing device is configured to determine the position of at least part of the handling device (103) and/or the load 105) with reference to a signal from the radar sensor (111). At least a part (405) of the handling device (103) is radar-absorbing.

A method and system of training a machine learning neural network (MLNN) in anatomical degenerative conditions in accordance with anatomical dynamics. The method comprises receiving, in a first input layer of the MLNN, from a millimeter wave (mmWave) radar sensing device, a first set of mmWave radar point cloud data representing a first gait characteristic of a subject in motion, comprising an arm swing velocity, receiving, in a second layer, a second set of mmWave radar point cloud data representing a second gait characteristic comprising a measure of dynamic postural stability, the input layers being interconnected with an output layer of the MLNN via an intermediate layer, and training a MLNN classifier in accordance with a classification that increases a correlation between a degenerative condition of the subject as generated at the output layer and the sets of mmWave point cloud data.

A method and apparatus with radar signal processing are included. A method includes transmitting, through transmission antenna elements, a radar signal at a transmission time interval corresponding to a time division multiplexing (TDM) latency, receiving a reflected signal of the radar signal through reception antenna elements, determining directions of arrival (DOAs) respectively corresponding to the transmission antenna elements by classifying radar data corresponding to the reflected signal, wherein the classifying is based on the transmission time interval, determining an unambiguous element of a phase error element by applying an ambiguous Doppler velocity that is based on the radar data to the phase error element of the individual DOA data, and determining integrated DOA data corresponding to the transmission antenna elements by integrating the individual DOA data by suppressing an ambiguous element of the phase error element.

Radar based HR and BR measurements by simultaneous decoding is a technical problem due to presence of intermodulation of BR and HR harmonics, which degrades simultaneous decoding. Embodiments herein provide a method and system for unobtrusive liveliness detection and monitoring of a subject using a Dual Frequency Radar (DFR) in an IOT network. The system has the capability to completely process the captured raw signals onboard to by applying required signal conditioning and extraction of relevant information using unique signal processing techniques for determining the HR and the BR of the subject accurately. The intermodulation of BR and HR harmonics is eliminated by the system by performing frequency spectrum averaging of both radars signals, which improves the accuracy. Further, the system is configured with a light MQTT protocol and encoding modules for any data to be shared for off board processing, ensuring data security and privacy compliance.

Aspects of the disclosed technology provide solutions for performing odometry and in particular, for performing odometry by filtering moving objects from a scene using sensor data. In some aspects, a process can include steps for receiving a first set of sensor data corresponding with a plurality of objects in a scene, determining one or more moving objects and one or more stationary objects from among the plurality of objects, and receiving a second set of sensor data. In some aspects, the process can further include steps for filtering the second set of sensor data to remove data associated with the one or more moving objects and generating odometry data associated with the filtered second set of sensor data. Systems and machine-readable media are also provided.

Various embodiments wirelessly detect micro gestures using multiple antenna of a gesture sensor device. At times, the gesture sensor device transmits multiple outgoing radio frequency (RF) signals, each outgoing RF signal transmitted via a respective antenna of the gesture sensor device. The outgoing RF signals are configured to help capture information that can be used to identify micro-gestures performed by a hand. The gesture sensor device captures incoming RF signals generated by the outgoing RF signals reflecting off of the hand, and then analyzes the incoming RF signals to identify the micro-gesture.

Methods for detecting radar targets are provided. According to one exemplary embodiment, the method includes providing a digital radar signal having a sequence of signal segments. Each signal segment of the sequence is respectively associated with a chirp of a transmitted RF radar signal. The method further includes detecting one or more radar targets based on a first subsequence of successive signal segments of the sequence. For each detected radar target, a distance value and a velocity value are determined. If a group of radar targets having overlapping signal components has been detected, a respective spectral value is calculated for each radar target of the group of radar targets based on a second subsequence of the sequence of signal segments and further based on the velocity values ascertained for the group of radar targets.