Sensor and Sensor Fusion Technology in Autonomous Vehicles: A Review

Area of Science:

• Robotics
• Computer Vision
• Sensor Technology

Background:

• Automated driving relies heavily on sensor perception for safety and feasibility.
• Accurate sensor calibration is crucial before sensor fusion and obstacle detection.
• Multiple integrated sensors enhance the reliability of autonomous systems.

Purpose of the Study:

• To evaluate the capabilities and technical performance of common autonomous vehicle sensors.
• To review sensor calibration techniques and open-source packages for multi-sensor systems.
• To summarize sensor fusion approaches and state-of-the-art algorithms for object detection.

Main Methods:

• Focus on vision cameras, LiDAR, and radar sensors under various operating conditions.
• Overview of three primary sensor calibration categories and open-source calibration tools.
• Summary of three main sensor fusion approaches and current multi-sensor fusion techniques.

Main Results:

• Provides an end-to-end review of hardware and software methods for sensor fusion object detection.
• Evaluates sensor performance and compatibility with commercial systems.
• Highlights challenges and proposes future research in sensor fusion for autonomous driving.

Conclusions:

• Sensor fusion object detection requires a comprehensive understanding of hardware and software integration.
• Accurate calibration and fusion are key to advancing autonomous driving safety.
• Further research is needed to address current challenges in sensor fusion technology.