To comply with the increasing complexity of new mechatronic systems and stricter safety regulations, advanced estimation algorithms are currently undergoing a transformation towards higher model complexity. However, more complex models often face issues regarding the observability and computational effort needed. Moreover, sensor selection is often still conducted pragmatically based on experience and convenience, whereas a more cost-effective approach would be to evaluate the sensor performance based on its effective estimation performance. In this work, a novel estimation and sensor selection approach is presented that is able to stabilise the estimator Riccati equation for unobservable and non-linear system models. This is possible when estimators only target some specific quantities of interest that do not necessarily depend on all system states. An Extended Kalman Filter-based estimation framework is proposed where the Riccati equation is projected onto an observable subspace based on a Singular Value Decomposition (SVD) of the Kalman observability matrix. Furthermore, a sensor selection methodology is proposed, which ranks the possible sensors according to their estimation performance, as evaluated by the error covariance of the quantities of interest. This allows evaluating the performance of a sensor set without the need for costly test campaigns. Finally, the proposed methods are evaluated on a numerical example, as well as an automotive experimental validation case.

Torsional vibrations play a critical role in the design and operation of a mechanical or mechatronic drivetrain due to their impact on lifetime, performance, and cost. A magnetic spring allows one to reduce these vibrations and improve the actuator performance yet introduces additional challenges on the identification. As a direct torque measurement is generally not favourable because of its intrusive nature, this paper proposes a nonintrusive approach to identify torsional load profiles. The approach combines a physics-based lumped parameter model of the torsional dynamics of the drivetrain with measurements coming from a motor encoder and two MEMS accelerometers in a combined state/input estimation, using an augmented extended Kalman filter (A-EKF). In order to allow a generic magnetic spring torque estimation, a random walk input model is used, where additionally the angle-dependent behaviour is exploited by constructing an angle-dependent estimate and variance map. Experimental validation leads to a significant reduction in bias in the load torque estimation for this approach, compared to conventional estimators. Moreover, this newly proposed approach significantly reduces the variance on the estimated states by exploiting the angle dependency. The proposed approach provides knowledge of the torsional vibrations in a nonintrusive way, without the need for an extensive magnetic spring torque identification. Further, the approach is applicable on any drivetrain with angle-dependent input torques.