Data-Driven Estimation of Quadrotor Motor Efficiency via Residual Minimization

TL;DR

The paper addresses online estimation of quadrotor motor efficiency by formulating a residual-minimization problem over trajectory data. It introduces a sliding-window, constrained nonlinear optimization solved with a primal-dual interior-point method, augmented by an outer Iteratively Reweighted Least Squares loop for robust outlier rejection. Key contributions include a well-defined trajectory residual formulation, a robust optimization algorithm with log-barrier constraints, and demonstrated robustness to degradation and faults while achieving similar long-term accuracy to an EKF baseline. Practically, this approach supports fault detection, health monitoring, and predictive maintenance in aerial robotics, and is amenable to onboard implementation and learning-based extensions.

Abstract

A data-driven framework is proposed for online estimation of quadrotor motor efficiency via residual minimization. The problem is formulated as a constrained nonlinear optimization that minimizes trajectory residuals between measured flight data and predictions generated by a quadrotor dynamics model. A sliding-window strategy enables online estimation, and the optimization is efficiently solved using an iteratively reweighted least squares (IRLS) scheme combined with a primal-dual interior-point method, with inequality constraints enforced through a logarithmic barrier function. Robust z-score weighting is employed to reject outliers, which is particularly effective in motor clipping scenarios where the proposed estimator exhibits smaller spikes than an EKF baseline. Compared to traditional filter-based approaches, the batch-mode formulation offers greater flexibility by selectively incorporating informative data segments. This structure is well-suited for onboard implementation, particularly for applications such as fault detection and isolation (FDI), health monitoring, and predictive maintenance in aerial robotic systems. Simulation results under various degradation scenarios demonstrate the accuracy and robustness of the proposed estimator.

Figures (8)

• Figure 1: Architecture of the proposed motor efficiency estimator integrated with geometric tracking controller.
• Figure 2: Comparison between the EKF baseline and the proposed estimator under gradual voltage-induced degradation with thrust noise.
• Figure 3: Comparison between the EKF baseline and the proposed estimator under abrupt motor fault injection with thrust noise.
• Figure 4: Comparison between the EKF baseline and the proposed estimator under combined degradation and abrupt fault scenarios with thrust noise.
• Figure 5: Root mean square error (RMSE) and standard deviation of motor efficiency estimation under various degradation and fault scenarios, comparing the EKF baseline with the proposed method.