News
Presented “Towards Resilient Tracking in Autonomous Vehicles: A Distributionally Robust Input and State Estimation Approach” at IFAC IAV 2025 (Intelligent Autonomous Vehicles Symposium) in May 2025.
This paper proposes a novel framework for Distributionally Robust Input and State Estimation (DRISE) for autonomous vehicles operating under model uncertainties and measurement outliers.
The proposed framework improves the input and state estimation (ISE) approach by integrating distributional robustness, enhancing the estimator’s resilience and robustness to:
- Adversarial inputs
- Unmodeled dynamics
- Measurement outliers
- Model uncertainties
Moment-based ambiguity sets capture probabilistic uncertainties in both:
- System dynamics
- Measurement noise
This offers:
- Analytical tractability
- Efficient handling of uncertainties in mean and covariance
- Minimization of worst-case estimation error
- Robustness against deviations from nominal distributions
The effectiveness of the proposed approach was validated through simulations conducted in the CARLA autonomous driving simulator, demonstrating:
- Improved performance in state estimation accuracy
- Enhanced robustness in dynamic and uncertain environments
- Better resilience to real-world driving scenarios