---
title: Error-State Kalman Filtering
url: https://www.emergentmind.com/topics/error-state-kalman-filtering
type: topic
---

# Error-State Kalman Filtering

Error-state Kalman filtering (ESKF) constitutes a principled approach for recursive state estimation of nonlinear systems, particularly in scenarios involving orientation and inertial navigation. Unlike classical Kalman filtering, ESKF estimates the evolution of the error between a nominal (predicted) trajectory and the true system state, enabling robust linearization and enhanced stability—especially for systems with complex rotational or multi-rate structure. ESKF variants, including classical (additive), invariant (multiplicative, Lie group-based), and high-order Taylor expansion-based forms, address the challenges of high-dynamic motion, sensor fusion, global- and body-frame switching, and real-time deployability [2507.13179], [2511.00453], [2310.01844]. 

## 1. System and Error-State Formulation

The ESKF paradigm decomposes the system state $x$ into a nominal trajectory $\bar{x}$ and a small error-state $\delta x$ such that $x \approx \bar{x} \boxplus \delta x$. For six-degree-of-freedom motion in $\mathbb{R}^3 \times S^3$ (with position, velocity, acceleration, quaternion, angular velocity, and angular acceleration), the error-state is defined as:
\[
\delta x = 
\begin{bmatrix}
\delta p \\
\delta v \\
\delta a \\
\delta \theta \\
\delta \omega \\
\delta \alpha
\end{bmatrix}
\in \mathbb{R}^{15},
\]
where $\delta \theta = 2\,\log(q \otimes \bar{q}^{-1})$ is the minimal $\mathrm{SO}(3)$ rotation vector.

For systems evolving on Lie groups (e.g., $SE_2(3)$ for full pose, $SO(3)$ for attitude), the error-state may be defined right- or left-invariantly: $\eta^{R} = \hat{\chi}^{-1} \chi = \exp(\xi^{R})$ (right-invariant) or $\eta^L = \hat{\chi} \chi^{-1} = \exp(\xi^L)$ (left-invariant), where $\chi$ collects all physical states as an element of the matrix Lie group and $\xi$ is the minimal vector error [2310.01844].

This separation enables all state updates to be performed via group operations or minimal-coordinate updates, and ensures that linearization remains accurate provided errors are small.

## 2. High-Order Error-State Dynamics and Discrete-Time Propagation

Classical ESKF employs a first-order Taylor expansion, yielding linearized error-state dynamics:
\[
\dot{\delta x} = A \delta x + G w,
\]
with $A = \left.\frac{\partial f}{\partial x}\right|_{\bar{x}}$ (Jacobian), $G$ process noise mapping, and $w$ white noise. High-order ESKF, as introduced in motion prediction for edge XR, systematically expands the dynamics as a Taylor series in $\delta x$ up to the third order:
\[
f(x) = f(\bar{x} + \delta x) \approx f(\bar{x}) + A\,\delta x + \frac{1}{2}H[\delta x \otimes \delta x] + \frac{1}{6}T[\delta x \otimes \delta x \otimes \delta x] + ...
\]
where $H$ and $T$ are the second and third derivative tensors, respectively [2507.13179]. The resulting discrete-time update for the $k$th prediction step employs series-truncated Magnus-Taylor expansions for the state transition and higher-order error-state terms (quantified as $\Psi^{(i)}$).

Covariance propagation follows:
\[
P_{k|k-1} = \Phi_k P_{k-1|k-1} \Phi_k^\top + Q_k + \textrm{(higher-order terms)},
\]
where $\Phi_k$ is the state transition, $Q_k$ the process-noise covariance, and higher-order corrections are typically truncated beyond the second or third order for computational tractability.

For right-invariant ESKF on Lie groups, the propagation is characterized by exponential maps:
\[
\Phi_k = \exp(F_t^R\,\Delta t), \quad 
Q_k = \int_0^{\Delta t} \Phi(\Delta t-\tau) G_t^R Q_m (G_t^R)^\top \Phi(\Delta t-\tau)^\top d\tau,
\]
enabling consistent scaling and invariance to large initial state errors [2310.01844].

## 3. Measurement Update and Covariance Transformation

Measurement updates in ESKF are linearized around the nominal state:
\[
z_k = h(x_k) + v_k, \qquad
\delta z \approx H_k\,\delta x + v_k,
\]
with $H_k$ the Jacobian of $h$ at $\bar{x}$. The innovation adopts a minimal, invariant representation, for example $y_k = z_k \ominus h(\bar{x}_{k|k-1}) = [p_\text{meas} - \bar{p}; 2 \log(q_\text{meas} \otimes \bar{q}^{-1})]$ for pose and orientation.

The Kalman gain $K_k$ and posterior covariance $P_{k|k}$ are then computed as per standard Kalman filter algorithm, and the correction $\delta x$ is "injected" back into the nominal state through a retraction (addition for $\mathbb{R}^n$, group composition for quaternions or general Lie group structure).

Covariance transformation-based ESKF (CT-ESKF) unifies error-state variants by mapping the covariance across different error-state definitions. Let $\xi_a = A(\hat{x})\,\xi_b$ relate two error coordinates with $P_a = A(\hat{x})\,P_b\,A(\hat{x})^\top$. After each update, one applies:
\[
P^+_{\mathrm{CT}} = T_{b\to a}\,P^+\,T_{b \to a}^\top, \quad T_{b \to a} = A(\hat{x}^+)^{-1}A(\hat{x}^-),
\]
to ensure the propagated covariance aligns with the target error-state basis, crucial for robust multi-sensor integration—especially in fusing global-frame (GNSS) and body-frame (ODO) observations [2511.00453].

## 4. Invariance Principles and Trajectory Independence

Invariant ESKF (InEKF) leverages the group-affine properties of propagation and observation models, yielding error-state dynamics and measurement Jacobians that are independent (or nearly so) of the estimated state trajectory. For $SE_2(3)$,
\[
F_t^R =
\begin{bmatrix}
F_{davp} & F_{davp2b} & 0 \\
0 & I & 0 \\
0 & 0 & F_{\alpha v}
\end{bmatrix},
\]
with most blocks state-independent except for entries related to non-group variables (e.g., biases) [2310.01844]. The result is improved stability, rapid convergence (e.g., $\sim$10 s after large attitude bias), and immunity to poor initialization as long as group-affine conditions are satisfied.

Analytical results show that, for propagation, EKF, L-InEKF, and R-InEKF all transmit the same information up to the coordinate basis, with differences arising only in Jacobian structure and their interaction with nonlinear measurement models. CT-ESKF enables switching between these bases without switching the core filter, preserving both consistency and information [2511.00453].

## 5. Implementation Complexity and Real-Time Performance

The computational trade-offs of ESKF frameworks are determined by the order of expansion, state-space dimension, and the use of efficient Lie group operations. High-order ESKF (e.g., PsudoESKF) with $n=15$ states carries $O(n^2)$ cost for Jacobian calculation, and $O(n^3)$ for series expansion or covariance updates. High-order tensors ($H$, $T$) induce $O(n^3)$ and $O(n^4)$ storage requirements, motivating selective truncation (quaternion blocks only or up to $i=3$ terms), the use of directional derivative approximations, or neglecting pure cross-terms to balance accuracy and compute demands [2507.13179].

Quaternion integration schemes (Zed12/Zed23) are employed to amortize costs of the exponential map, implicitly incorporating third-order corrections.

Empirical benchmarks confirm that per-step latency (sub–1 ms on Apple M1 CPUs) remains well below 10 ms sampling intervals, supporting real-time deployment for high-rate edge XR applications.

## 6. Comparative Performance and Applications

Extensive empirical studies validate ESKF performance in diverse application settings:

- **Edge XR motion prediction:** High-order ESKF (PsudoESKF) with third-order expansion achieves a reduction in position error (100 ms horizon, hard motion) from $\sim$54.1 mm (KF) to $\sim$17.8 mm (67% reduction), and orientation error from $3.39^\circ$ (KF) to $1.71^\circ$ (50% reduction). Under 50% packet loss, orientation error is reduced by 49.6% vs. KF [2507.13179].
- **Multi-sensor navigation:** CT-EKF achieves attitude RMSE better than EKF, L-InEKF, or R-InEKF (sub-degree accuracy), even under large initial attitude error or slow IMU propagation (e.g., 2 Hz) [2511.00453].
- **UAV full-state estimation:** The ES-RIEKF delivers 10% improvement in attitude MAE/RMSE and 4% improvement in position RMSE compared to standard ES-EKF; robustness is maintained during extended GNSS outages (maximum position error within 30 m over a 130 s GNSS denial) [2310.01844].

A plausible implication is that targeted selection of error basis and order of expansion, coupled with efficient Lie group computation, is essential to obtain robust, accurate, and real-time-capable Kalman filtering for sensor fusion tasks spanning robotics, inertial navigation, and edge compute vision systems.

## 7. Extensions and Synthesis with Learned or Aerodynamic Models

Contemporary approaches incorporate model-aided and data-driven augmentations, e.g., LSTM networks for drift-free prediction of aerodynamic angles (angle of attack/side-slip) using control-surface deflections and IMU data. These learned predictions integrate as pseudo-measurements within the invariant ESKF framework, reducing reliance on external (e.g., GNSS) sensors, and preserving estimation consistency under denial or degradation [2310.01844]. 

Moreover, entropy-based or confidence-driven classifiers can be integrated for motion predictability assessment, enabling dynamic adjustment of filter order, detection of low-predictivity regimes, and informed switching between computationally intensive and lightweight variants [2507.13179]. This convergence of model-based ESKF and adaptive or learning-aided components suggests a general trend toward hybrid architectures for resilient, high-accuracy navigation and state estimation.

---

**Key references**:  
- "Predictability-Aware Motion Prediction for Edge XR via High-Order Error-State Kalman Filtering" [2507.13179]  
- "CT-ESKF: A General Framework of Covariance Transformation-Based Error-State Kalman Filter" [2511.00453]  
- "Semi-Aerodynamic Model Aided Invariant Kalman Filtering for UAV Full-State Estimation" [2310.01844]

Source: https://www.emergentmind.com/topics/error-state-kalman-filtering