Blind Spot Detection Systems in Smart Vehicles: A Comprehensive Technical Review of Technologies, Methodologies, and Performance Analysis

Abstract

With the continuous rise in road traffic, blind spot-related accidents have become a persistent and serious threat to road safety, particularly involving large vehicles such as buses and trucks. Due to their size, these vehicles have significantly larger blind spots compared to passenger cars, leading to more severe and often fatal collisions, especially with vulnerable road users (VRUs) such as cyclists and pedestrians. As a result, there has been increasing focus on the development of Blind Spot Monitoring (BSM) systems, a critical component of Advanced Driver Assistance Systems (ADAS). These systems are designed to detect objects or vehicles in areas not directly visible to the driver, using technologies such as radar, ultrasonic sensors, and camera-based vision systems. Recent advances have incorporated deep learning algorithms and sensor fusion techniques to enhance detection accuracy and system responsiveness.

This review paper presents a comprehensive review of both traditional and modern blind spot detection methods, evaluating their effectiveness, limitations, and real-world applicability. In addition, this paper highlights a proposed vision-based approach using rear-view cameras, offering a low-cost yet reliable solution for blind spot detection in commercial vehicles. The study emphasizes the need for improved system integration, real-time performance and bidirectional alert mechanisms to prevent collisions and enhance overall road safety.