International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056 Volume: 07 Issue: 07 | July 2020 www.irjet.net p-ISSN: 2395-0072 © 2020, IRJET | Impact Factor value: 7.529 | ISO 9001:2008 Certified Journal | Page 1073 Multi-Sensor Fusion and Sensor Calibration for Autonomous Vehicles Smitha Gogineni Staff Engineer Instrumentation & Controls, Texas, USA ---------------------------------------------------------------------***---------------------------------------------------------------------- Abstract - Usage of Autonomous vehicles have grown in the recent years even though their safety records are still in question. While the manufacturers are adding more sophisticated sensors and technology to make them safer, there are still issues with cars classification and detection of objects around them. Automated sensors with the combination of automotive software and computers perform a vital role in autonomous driving as they monitor surroundings, detect obstacles, they allow the automation system to take over full control of the vehicle, thereby saving drivers a significant amount of time by doing tasks in much more efficient and safe ways and safely plan the routes and paths autonomously. While autonomous vehicle technology appears to be developing at a continual pace, so far no commercially available vehicles have yet passed the required level 4 ranking for road-safe autonomous vehicles as the contemporary autonomous vehicle (AV) systems face critical obstacles along the road to reaching the primary safety and reliability goals. There is still a huge amount of technology improvement that needs to be taken in order to ensure autonomous vehicle safety on the roads. This paper presents the current advancements of the autonomous vehicle driving technologies and points to the still existing performance challenges for the development of level 5 fully automated autonomous driving. Key Words: Autonomous vehicles, Advanced driver assistance systems, Autonomous driving, Automotive, Intelligent vehicles, LIDAR, Sensor Calibration, Sensor Fusion 1. INTRODUCTION This year is supposed to be a remarkable year for self- driving cars where all major autonomous vehicle (AV) car makers have boldly declared years ago that this year the full automation autonomous driving and the permanent backseat driver status will be achieved. Even with extraordinary efforts from many of the leading auto makers, the fully autonomous cars are still out of reach and almost every one of the above predictions has been rolled back as the engineering teams of all these companies realized the complexity of the target and that this is going to be a much more incremental process. Possibly the major technical hindrance, is adapting human intelligence that enables car driving which was taken for granted to be easily replicated to autonomous driving systems proving previous predictions to be far too optimistic. There is an imminent gap, an important fact that current levels of vehicle autonomy are adequately low and the accountability for supervisory actions still resides with the drivers to operate the vehicles safely. There are welfare benefits of autonomous vehicles that could possibly eliminate emissions, increase traffic efficiency, improve road safety with accurate driving decisional problems associated with the human infirmities of fatigue, misperceptions, distractions and intoxication in the context of driving. As such there is a strong need to develop the autonomous vehicles to SAE level 5 to reap the outcome of this autonomous revolution. 2. AUTONOMOUS VEHICLE SENSORS The classifications of autonomous driving are the adopted standards J3016 of the international engineering and automotive industry association, Society of Automotive Engineers SAE and U.S. Department of Transportation’s National Highway Traffic Safety Administration (NHTSA) are as follows Level 0: Driver only: the human driver controls everything independently, steering, throttle, brakes, etc. Level 1: Assisted driving: assistance systems help during vehicle operation (Cruise Control, ACC) – Year 2000 Level 2: Partial automation: the operator must monitor the system at all times. At least one system, such as cruise control and lane centering, is fully automated – Year 2013 Level 3: Conditional automation: the operator monitors the system and can intervene when necessary. Safety-critical functions, under certain circumstances, are shifted to the vehicle – Year 2018 Level 4: High automation: there is no monitoring by the driver required. Vehicles are designed to operate safety- critical functions and monitor road conditions for an entire trip. However, the functions do not cover all every driving scenario and are limited to the operational design of the vehicle – Year 2024 Level 5: Full automation: operator-free driving – Year 2030 2.1 Cameras The most commonly used and primary sensors by all top and leading driverless technology developers are video cameras, radar sensors, ultrasonic sensors and lidar sensors. These sensors are further classified as active and passive, while active sensors send out energy in the form of a wave and detect objects based upon the information received such as radar sensors, the passive sensors simply receive information from the environment without emitting a wave, such as cameras. Camera/Image Sensors Cars manufactured from 2018 already have the reverse cameras and the front cameras for lane departure warning