Gm Cruise Holdings LLC

Daniel Flores Tapia

Devin Cass

Anton Mario Bongio Karrman

Jack Stepanian

Kotung Lin

Jace Artichoker

The present disclosure is directed to the transmission and reception of sets of very high frequency electromagnetic (EM) signals in ways that allow a sensing apparatus to discriminate between different sets of transmitted EM signals. Here a sensing apparatus may sequentially transmit different sets of EM signals. Each of these different sets of signals may include an encoded identifier that uniquely identifies each respective signal set of the different sets of signals. Each of these signal sets may include several pulses of a particular frequency with a same relative phase relationship followed by pulses that have a different phase relationship. These changes in phase may be used to encode the unique identifiers into the different sets of transmitted EM energy and these identifiers may be used by a sensing apparatus to associate specific received sets of EM energy with specific sets of transmitted EM energy.

Frequency modulated continuous wave (FMCW) waveform generator/decoder

High-resolution point cloud formation based on the use of a point spread function kernel is disclosed. The proposed disclosure provides the means to generate high angular resolution point clouds by performing a matching process of the point spread function (PSF) to a series of adaptive hypotheses to accurately identify the target signatures using the maximum amount of available information. The proposed method comprises a two-stage processing chain. In the first step, the spatial and the Doppler spectrum of the target responses is calculated and decomposed in terms of the sensor point spread function (PSF). In the second step, the target PSF representation is then processed using a support vector machine to determine the presence of closely located targets that would not be possible to detect using typical automotive RADAR signal processing chains.

High-resolution point cloud formation in automotive-grade radar signals

The present technology determines noise in an image of a representation of an environment around an autonomous vehicle (AV). A sensor system receives a first image that represents an environment around the AV including at least one object that may be obscured by environmental conditions. The first image is analyzed to determine if pixels in the first image represent noise from the environmental conditions in comparison to other pixels in the first image that represents a higher degree of image detail. A second image can be generated using the pixels in the first image that represent the higher degree of image detail to result in an output that is less affected by noise from environmental conditions and may result in better performance by downstream systems relying of date from the sensor system.

Reducing environmental interference from images

Disclosed are systems and methods for scintillation-based neural network for RADAR target classification. In some aspects, a method includes generating a point cloud from radio frequency (RF) scene responses received from a radio detection and ranging (RADAR) sensor for a scanned scene; populating a rolling buffer with frame data from the point cloud, the frame data including radar cross section (RCS) values, RCS scintillation measurements, and velocity values for objects in the scanned scene; inputting the RCS scintillation measurements and velocity values for an object of the objects to a convolutional neural network (CNN); and receiving a classification of the object from the CNN, wherein the CNN is to utilize a probability density function (PDF) estimate of the RCS scintillation measurements and the velocity values to determine fits with one or more reference PDFs based on a Neyman-Pearson evaluation, and wherein the fits are assessed to classify the object.

Scintillation-based neural network for radar target classification

Methods and systems are provided for generating an on-demand distributed aperture by mechanical articulation. In some aspects, a process can include steps for determining a location of an autonomous vehicle, determining whether a maneuver requires long range detections or medium range detections based on the location of the autonomous vehicle, positioning at least two articulated radars based on the determining of whether the maneuver requires long range detections or medium range detections, and enabling a mode of resolution based on the positioning of the at least two articulated radars and by utilizing a static radar. Systems and machine-readable media are also provided.

Radar system for generating an on-demand distributed aperture by mechanical articulation

A radio detection and ranging (RADAR) sensor system signal generator is disclosed. The signal generator includes a lower frequency oscillator circuit to generate a first reference signal having a first frequency, a higher frequency oscillator circuit to generate a second reference signal having a second frequency that is higher than the first frequency, a set of frequency multipliers and signal mixers coupled to receive the first reference signal and the second reference signal, the set of frequency multipliers and signal mixers to generate a plurality of baseline signals, voltage-to-frequency converters to generate radar frequency waveforms at a plurality of specified frequencies according to chirp parameters for a plurality of frequency bands, and modulating radio frequency mixers to mix the plurality of baseline signals with the radar frequency waveforms at the plurality of specified frequencies to generate a multifrequency wavelet RADAR waveform.

Fractalet radar waveform generators

Multi-sensor radar microDoppler holography is described. An example of a method includes performing a 2D Fourier transform calculation for multiple sets of raw Radar sensor data to generate Doppler spatial frequency map data, the raw sensor data including at least a first set of raw sensor data associated with a first Radar sensor at a first location and a second set of raw sensor data associated with a second Radar sensor at a second, different location; performing a DWT calculation of the Doppler spatial frequency map data; applying phase compensation to generate phase compensated data transforms; adding the phase compensated data transforms to generate a sum; performing an inverse 2D Fourier transform calculation to generate a transform result; and extracting microDoppler information and Doppler information extraction from the transform result.

Multi-sensor radar microdoppler holography

For some embodiments of the present disclosure, systems and methods for using computer vision to guide processing of receive responses of RADAR sensors of a vehicle are described. A computer implemented method comprises receiving, with one or more receivers of RADAR sensors, environment receive RF responses, receiving region of interest (ROI) information including segments of a tentative set of ROIs with free space area boundaries from an image processor of an image processing chain for the computer vision, dynamically determining a number of polyphase filters based on a number of ROIs having dynamic scenes that are provided by the image processing chain, and adjusting parameters of the polyphase filters of a RADAR processing chain based on the ROI information.

Systems and methods for using computer vision to guide processing of receive responses of radar sensors of a vehicle

A radar apparatus at a vehicle may transmit a set of radar energy that is reflected off an object after which the reflected radar energy may be received at the radar apparatus. When relative motion of a radar apparatus and an object include movement in two different directions, power of the reflected radar signals may be spread out in a manner that makes data associated with the reflected radar signals difficult to interpret. This spreading out of the radar signals can make mappings of the received radar data appear to be smeared or distorted. To compensate for this smearing effect, mappings of this smeared radar data may be compared with curves from which compensation factors may be identified. These compensation factors may allow a processor to perform calculations to generate updated mappings of the radar data that may allow the processor to more accurately identify characteristics of the object.

High-resolution radar target signature smearing correction

Architectures and techniques for radar interference detection are provided. A radar sensor system in accordance with the present disclosure may receive, via a radio frequency (RF) receiver, radar signals including a radar signal of interest and one or more interfering radar signals. The radar sensor system may calculate a Doppler spectrum for each of the radar signals and perform a chirplet transform on the Doppler spectrum to generate various waveform parameters. A Principal Component Analysis (PCA) may be performed on the waveform parameters to extract frequency features of the radar signals. The radar sensor system may classify the frequency features using a classifier to identify interfering frequency features associated with the interfering radar signals using a classifier. The radar sensor system may further extract interfering waveform information based on the interfering frequency features of the interfering RF signals. Interference mitigation may be performed utilizing the interfering waveform information.

Radar interference detection and mitigation

The disclosed technology provides solutions for improving perception systems and in particular for improving object identification based on sensor point cloud data, such as radar point cloud data. A process of the disclosed technology can include steps for receiving point cloud data comprising a plurality of radar points, wherein each of the radar points corresponds with a first object or a second object in an environment, generating a semantic label for each of the radar points, and clustering the plurality of radar points based on the semantic label for each of the radar points, to generate a first point cloud cluster for the first object and a second point cloud cluster for the second object. Systems and machine-readable media are also provided.

Semantic segmentation based clustering

Methods and apparatus disclosed within provide a solution to problems associated with the use of either high frequency or low frequency radar signals. Methods of the present disclosure may transmit high frequency radar signals, transmit low frequency radar signals, receive reflected radar signals, and process the received radar signals using parameters respectively suited for processing high frequency and low frequency radar signals. Evaluations may be performed that allow an apparatus to adapt for limitations associated with the processing of high frequency radar data, the processing of low frequency radar data, or both. Different correlation functions may be performed that allow the apparatus to identify objects and to identify object velocities using different sets of program code instructions. These different evaluations and correlations may result in the generation of a set of 'point-cloud' information that may be used by other processes of a sensing apparatus.

Multiple frequency fusion for enhanced point cloud formation

Processing of a fractalet radio detection and ranging (RADAR) signal is described. A reference fractalet waveform is received. The fractalet waveform includes self-similar waveforms having lower frequency bands and frequency bands. A reflected fractalet waveform received via one or more antennae is decoded. A waveform profile of chirplet transforms of signals in the lower frequency bands within the reflected fractalet waveform are compared to the reference fractalet waveform. Time spans corresponding to the subset of lower frequency bands are determined. Signals from the higher frequency bands are extracted from the reflected fractalet waveform. Chirplet transforms for the extracted signals from the higher frequency bands are determined for the determined time spans. Spatial frequency components along azimuth direction and elevation directions are calculated for targets based on the chirplet transforms for the extracted signals from the higher frequency bands.

Fractalet radar processing

Techniques and architectures for managing radar sensor processing chains. A first high-frequency radio signal is received with a first RF receiver in the plurality of RF sensor suites on a host platform. The received high-frequency radio signal is converted to a lower second frequency range. A chirplet transform is performed on the signal in the second frequency range. Stored relative location information for a second RF receiver in the plurality of RF sensor suites is retrieved. Radar waveform information corresponding to the second RF receiver in a processing stream corresponding to the first RF receiver is extracted by utilizing the retrieved information and results from the chirplet transform. A point cloud is generated based on the converted signal in the second frequency range and the extracted radar waveform information.

Radar sensor processing chain

Architectures and techniques for radar signaling in emergency scenarios. A high-frequency radio signal in a first frequency range from a remote device with a local radio frequency (RF) receiver. The received radio signal in the first frequency range is converted to a corresponding signal in a second and lower frequency range. Signal phase information in the lower frequency signal is modified to generate a modified signal in the lower frequency range. The modified signal in the lower frequency range is converted to the first frequency range. The modified signal in the first frequency range is transmitted to the remote device with an RF transmitter.

Growing community of inventors

Fairfield, CA, United States of America

Daniel Flores Tapia