Placement Method of Multiple Lidars for Roadside Infrastructure in Urban Environments.

Area of Science:

• Robotics and Intelligent Systems
• Computer Vision
• Transportation Engineering

Background:

• Autonomous vehicles (AVs) face sensor limitations due to physical constraints.
• Infrastructure-based sensors can augment AV sensing capabilities through wireless communication.
• Optimal placement of roadside sensors is critical to reduce blind spots and maximize coverage.

Purpose of the Study:

• To propose a novel method for optimizing the placement of multiple sensor systems within road infrastructure.
• To address the challenge of sensor placement for enhanced autonomous vehicle perception.

Main Methods:

• Formulation of the sensor placement problem using a voxel grid representation.
• Application of a genetic algorithm to solve the optimization challenge and identify optimal sensor locations.
• Experimental validation using lidar sensor data.

Main Results:

• The proposed method effectively identifies optimal positions for roadside sensors.
• Demonstrated reduction in blind spots and improvement in sensor coverage area.
• Validation of the approach's efficacy through real-world lidar sensor experiments.

Conclusions:

• The genetic algorithm-based voxel grid approach provides an effective solution for roadside sensor placement.
• This method enhances the sensing capabilities of autonomous vehicles by overcoming inherent sensor limitations.
• Optimized sensor deployment is crucial for improving the safety and reliability of autonomous driving systems.