Experimental Analysis of Various Blockage Performance for LiDAR Sensor Cleaning Evaluation.

Area of Science:

• Engineering
• Robotics
• Sensor Technology

Background:

• Autonomous driving relies on sensors like cameras, LiDAR, and radar.
• Environmental contaminants (dust, bird droppings, insects) degrade sensor performance.
• Limited research exists on sensor cleaning technologies.

Purpose of the Study:

• To evaluate sensor cleaning rates under various conditions.
• To identify critical factors affecting cleaning effectiveness.
• To compare cleaning performance against different types of contaminants.

Main Methods:

• Testing various blockage types (dust, bird droppings, insects) and concentrations.
• Evaluating cleaning using washer (0.5 bar/s) and air (2 bar/s) on a LiDAR window.
• Assessing the impact of dryness on cleaning efficiency.

Main Results:

• Blockage type, concentration, and dryness were identified as the most significant factors in cleaning effectiveness.
• Performance of cleaning methods was evaluated against standard dust and novel contaminants.
• Specific washing and air-jet parameters were tested for optimal results.

Conclusions:

• The study provides a framework for evaluating sensor cleaning methods.
• Results can ensure the reliability and economic feasibility of sensor cleaning systems.
• Understanding blockage characteristics is vital for effective autonomous vehicle sensor maintenance.