SPRADD2 august   2023 AM62A3 , AM62A3-Q1 , AM62A7 , AM62A7-Q1

 

  1.   1
  2.   Abstract
  3.   Trademarks
  4. 1 Introduction
  5. 2 AM62A Processor
  6. 3 System Block Diagram
  7. 4 Driver and Occupancy Mirror System Data Flow
  8. 5 Deep Learning Acceleration
  9. 6 Functional Safety in DMS/OMS Applications Using AM62A
    1. 6.1 Overview of Functional Safety Features on AM62A
  10. 7 Functional Safety Targets and Assumptions of Use
  11. 8 Functional Safety in DMS/OMS Data Flow
  12. 9 LED Driver Illumination Use Case
  13. 10Summary
  14. 11References

1 Introduction

New Car Assessment Program (NCAP) regulations are a set of standards and testing procedures used to evaluate the safety performance of new vehicles. NCAP programs exist in all regions around the world including the United States, Europe, Australia and Japan, among others. Driver monitoring systems and occupancy monitoring systems play a critical role in NCAP regulation compliance by enhancing vehicle safety and occupant protection and directly impact the safety rating of the vehicle.

Driver monitoring systems (DMS) monitor the driver’s attention, alertness and behavior to ensure they are actively engaged in the driving task. These systems typically deploy camera based and computer vision algorithms to detect signs of driver distraction, drowsiness, or impairment. DMS systems can also be integrated with other vehicle safety features such as lane departure warning and adaptive cruise control to improve their effectiveness.

Occupancy monitoring systems (OMS) are designed to detect the presence of vehicle occupants, including passengers and, in some cases, pets. These systems are generally also camera based and deep learning algorithms are used to determine whether seat belts are being used, if there are occupants in specific seats (for airbag deployment) and if the occupants meet certain requirements (such as child seats for children). OMS systems can help prevent or minimize injuries by providing seat belt reminders or adjusting airbag deployment based on the weight or size of the occupants. These systems are especially relevant for evaluating child occupant protection and can contribute to the safety rating of a vehicle.

This document provides an overview of the AM62A processor. Next, it is shown how its heterogeneous architecture enables deployment of DMS and OMS algorithms with the lowest compute and power. A typical DMS/OMS dataflow is presented and how the hardware feature sets in AM62A are used in a DMS/OMS system are discussed. This paper also highlights innovations in hardware like the ability to process RGB-IR sensors that enable the ability to create a complete DMS/OMS system with a single camera. A critical aspect in any automotive system is functional safety and we show the different functional safety considerations that need to be taken into account when building a DMS/OMS system with AM62A.