Hybrid Fusion of RADAR, Range-Finder LiDAR, and Ultrasonic Level Meter Sensors for Linear Sensing Applications

Collision avoidance systems serve as a major aspect of autonomous vehicles, while sensors form a crucial section of collision avoidance systems. The challenge of assembling sensors that can perceive the environment in a manner similar to human senses is being continuously explored by researchers to enhance the abilities of collision avoidance systems; more affordable and simpler methods are also being considered to reduce their complexity. This paper considers sensor fusion of radio detection and ranging (RADAR), Range finder light detection and ranging, and Ultrasonic sensors for a collision avoidance system. The concept of this paper is to exclude vision sensors from the fusion method. While vision sensors have been considered a fundamental sensor for collision avoidance systems by many researchers, the flaws of vision sensors when detecting transparent objects or under poor visibility conditions cannot be ignored. At distances below 1 m, the RADAR sensor will be rendered inactive, while at distances above 15 m, the ultrasonic sensors will be rendered inactive. All sensors are active at distances between 1 m and 15 m. The fusion method in this paper establishes an algorithm for sensor-fusion of the three sensors based on their linear accuracies.