Ineda CRoME Technology powering next generation AI systems
AI IN THE DRIVER SEAT FOR AUTONOMOUS DRIVING (AD)
The automotive industry is going through a revolution with ADAS (Autonomous Driver Assist Systems) getting deployed in cars and AD (Autonomous Driving) getting ready for deployment over the next few years. Artificial Intelligence is the key force behind ADAS and AD. Ineda’s unique System and Silicon architecture enables a flexible system to either process data at the edge or at the center and thus harnessing the power of ML/DNN/CNN/RNN and DRL to realise Level-5 AD Systems. Ineda leads application of AI as per the customer needs and customizes the silicon, algorithms and system software as per various use cases.
Role of AI in the evolution of Level 5 Compliant AD Technology
Machine Learning Use case Example 1 : Radar based Object Classification
Many of the ADAS functions like LCA, BSD, FCW, EBA, ACC, etc can enhance driving safety by classifying the objects encountered around the car. Ineda is engaged in developing the object classifier that can classify tens to hundreds of objects. We utilize state of the art ML/DL algorithms, data visualization techniques, and feature engineering to achieve highly accurate object classification.
Machine Learning use case example 2 : USRR Radar based Gesture Recognition
Touchless hand gesture recognition systems are becoming important in automotive user interfaces as they improve safety and comfort of the driver. Most vision-based gesture recognition systems available in the market have been developed for environments with controlled illumination. The interior of a car is a challenging environment because the lighting conditions vary a lot. Compared to colour and depth sensors, radars are robust to ambient illumination conditions, have lower cost and computational complexity, and use less power. Unique micro-Doppler frequency modulation signatures are produced by different hand gestures. These signatures are decoded with short-range radar sensor mounted on the car dash-board. We, at Ineda, use AI techniques to decode these signatures.
Machine Learning Use case Example 3 : Multi-sensor fusion based preventive maintenance
Ineda’s CRoME technology enables systems that use sensors to capture vehicle sound and vibration signals and analyze it for potential future failures of auto-parts such as air-filter, battery, air conditioning etc. This preventive maintenance system uses ML techniques to sense the degradation in the quality of the parts of the car and provides alerts to the user.