Intelligent Active Suspension Control Method Based on Hierarchical Multi-Sensor Perception Fusion

Area of Science:

• Automotive Engineering
• Control Systems
• Robotics and Intelligent Systems

Background:

• Intelligent suspension systems are crucial for vehicle dynamics, ride comfort, and safety.
• Current systems often face challenges in precise control due to limited sensor data.
• Sensor fusion offers a pathway to integrate diverse data for enhanced performance.

Purpose of the Study:

• To develop and validate a hierarchical multi-sensor fusion framework for active suspension control.
• To improve control precision and adaptivity to different road conditions.
• To enhance vehicle dynamic stability, ride comfort, and occupant safety.

Main Methods:

• Utilized a binocular vision system for detecting road features (lane curvature, speed bumps) and measuring distances.
• Integrated Global Positioning System (GPS) and inertial measurement unit (IMU) data for predicting road elevation profiles.
• Implemented a BP-PID control strategy with mode-switching rules and optimized suspension parameters using ant colony optimization.
• Validated the framework using a hardware-in-the-loop (HIL) simulation platform.

Main Results:

• Demonstrated significant reductions in vertical acceleration (5.37%), pitch acceleration (9.63%), and roll acceleration (11.58%).
• The proposed framework effectively adapted to flat, curved, and obstacle-laden road conditions.
• Hardware-in-the-loop simulations confirmed the system's efficacy and robustness.

Conclusions:

• The hierarchical multi-sensor fusion framework provides a robust and effective solution for active suspension control.
• This approach enhances vehicle dynamic stability, ride comfort, and safety through precise, adaptive control.
• The integration of vision, GPS, and IMU data represents a significant advancement in intelligent vehicle systems.