Trajectory Tracking Using Cumulative Risk-Sensitive Finite Impulse Response Filters.

Area of Science:

• Robotics and Control Systems
• Signal Processing

Background:

• Trajectory tracking is essential for autonomous driving and robotic motion control.
• Existing Infinite Impulse Response (IIR) filters face limitations in complex, dynamic environments due to model uncertainties.

Purpose of the Study:

• To propose a novel robust finite impulse response (FIR) filter for enhanced trajectory tracking accuracy and robustness.
• To address limitations of IIR filters by integrating a cumulative risk-sensitive criterion into an FIR structure.

Main Methods:

• Development of a novel robust finite impulse response (FIR) filter.
• Integration of a cumulative risk-sensitive criterion with the FIR filter structure.
• Validation through comprehensive vehicle trajectory tracking experiments.

Main Results:

• The proposed FIR filter significantly reduces average tracking error compared to Kalman Filter (KF), Risk-Sensitive Filter (RSF), and Unbiased FIR (UFIR) filter.
• Demonstrated superior robustness in complex and uncertain environmental scenarios.
• Effective mitigation of model mismatches and temporary modeling uncertainties.

Conclusions:

• The novel robust FIR filter offers an effective solution for trajectory tracking applications.
• The filter shows high suitability for dynamic and uncertain environments common in autonomous systems.
• This work has broad potential for practical implementation in autonomous driving and robotics.