Estimation

Estimation reconstructs latent vehicle state from noisy, asynchronous sensors under model uncertainty. This chapter focuses on Bayesian and complementary estimators used by downstream control and planning layers.

Core Questions

• How should process and measurement uncertainty be represented?
• When does linearization error dominate filter performance?
• What is the observability footprint of each sensor suite?

Algorithms

• Complementary Filter
• Extended Kalman Filter
• Unscented Kalman Filter
• GPS-IMU Fusion
• Particle Filter

Prerequisites

• Continuous-time rigid body dynamics
• Gaussian estimation and covariance propagation
• IMU and GPS measurement models