The automotive RADAR market is growing at a substantial rate. The key factor driving the growth of the automotive radar market is the rising requirements for vehicles to be equipped with ADAS and Safety features. These features reduce road accidents thus enhancing the safety of passengers as well as pedestrians. In addition to ADAS, Radar technology is also used in in-cabin monitoring as well as smart city applications.
SoCtronics has experience on a wide range of Radar based algorithms such as Object detection and tracking, In-cabin Vital signs monitoring, Hand gesture recognition, In-cabin occupancy detection, Traffic vehicle monitoring etc. TI’s mmWave radar sensors AWR1243, AWR1443 and AWR1642 are used for development and real time verification of the algorithms.
The following services are offered by SoCtronics for radar systems and chipset manufacturers:
- . Design, develop and customize Radar based DSP and AI algorithms based on use-case
- . Port Radar DSP and Radar based classification (ML/Deep learning) algorithms from MATLAB or Python to C for Embedded platforms, for Radar edge processing
- . Optimize MIPS and memory according to the resources available on Embedded platform.
- . Tune algorithms to work for different Radar sensors.
Soctronics is an ideal partner for Radar algorithm design and implementation for automotive radar manufacturers.
Radar SDK for Object detection and tracking
Case-study
View more
Radar SDK from SoCtronics is a ready to use Signal processing pipeline for real time detection and tracking of the objects. The SDK includes high performance Signal processing algorithms for precise measurement of the target location.
The Signal processing pipeline comprises CFAR, DoA, Clustering, Object detection and tracking components. Each of the component in the pipeline is implemented in a modular fashion with well defined APIs. Based on the requirement, customers can seamlessly integrate entire SDK or some components into their application. The SDK is implemented on a Radar edge processor which supports Radar DSP hardware accelerators.
Radar based target classification using CNN
Case-study
View more
This solution is implemented for classifying traffic at traffic signal junction through radar mounted on pole. The objective is to classify the vehicles into different classes- 2 wheeler, 3-wheeler, 4-wheeler, pedestrian. The implementation uses CNN with multiple Convolution and Fully connected layers.
The Radar mounted on traffic poles will receive the back scattered signal from different vehicles. Signal processing techniques are used to filter out the static objects, identify the regions where the targets are present and extract the features of the target. In training mode, labeled features are used for training the model. In real time, the trained model is used for classifying the vehicles based on the input features. For all the 4 classes, the accuracy of prediction is more than 90%. This accuracy is achieved by carefully selecting the chirp configuration, selecting the correct features which distinguish the targets, selecting the best CNN algorithm/network and training and tuning the model.
Radar based In-cabin Vital signs monitoring
Case-study
View more
Vital Signs (Heart Rate ,Breath Rate ) in the Car, are continuously detected using Radar with high accuracy for alerting in case of any medical emergency.
The algorithm detects vital signs of static personnel sitting in front of the Radar both single and multiple targets. The Radar signal reflected from the target consists of phase variations due to chest wall displacement caused by Breathing and Heart beat. The signal also consists of reflections from static objects which is clutter. A Novel technique is used to remove the static clutter and process only the Micro-Doppler signatures from the targets. Signal processing methods are used to analyze the Micro-Doppler signatures and extract Vital signs accurately.
