Bearing load estimation would form a valuable addition to the fields of condition monitoring and system control. Despite effort spend on its development by all major bearing manufacturers no product solution has come to market yet. This can be attributed to both the complexity in conditioning of the strain measurement as well as its non-linearity with respect to the bearing loading. To address these issues, this paper proposes a novel approach based on modeling of the physical behavior of the bearing. An Extended Kalman Filter including a novel strain model is applied for signal conditioning whereas an Unscented Kalman Filter including a semi-analytical bearing model is proposed for reconstruction of the bearing load. An experimental study in both laboratory and field conditions shows that the proposed cascaded Kalman filtering approach leads to accurate estimates for all four considered bearings loads in various loading conditions. Besides an improvement on accuracy, the novel approach leads to a reduction in calibration effort.@en

This paper proposes a novel semi-analytical bearing model addressing flexibility of the bearing outer race structure. It furthermore presents the application of this model in a bearing load condition monitoring approach. The bearing model is developed as current computational low cost bearing models fail to provide an accurate description of the more and more common flexible size and weight optimized bearing designs due to their assumptions of rigidity. In the proposed bearing model raceway flexibility is described by the use of static deformation shapes. The excitation of the deformation shapes is calculated based on the modelled rolling element loads and a Fourier series based compliance approximation. The resulting model is computational low cost and provides an accurate description of the rolling element loads for flexible outer raceway structures. The latter is validated by a simulation-based comparison study with a well-established bearing simulation software tool. An experimental study finally shows the potential of the proposed model in a bearing load monitoring approach.

@en

Inertial or motion cue plays a significant role for the achievement of an immersive feeling in driving simulators. The control loop is responsible for mimicking as close as possible the real vehicle accelerations as the inertial feedback to the driver, while keeping the motion system within its physical limitations. Several algorithms have been designed for this purpose, seeking for the optimal compromise between realistic accelerations fidelity and actuation dynamics restrictions. Motion cueing algorithms can hence vary depending on the dynamics of the maneuver, computational efficiency required and motion system configuration. Several algorithms can be found in literature, from the classical approach, based on a combination of high and low-pass filters on the vehicle accelerations, to more recent strategies based on Model Predictive Control (MPC). Assessing the success of one solution is a complex task that would involve the prototype implementation on hardware as well as the availability of several test subjects. Nevertheless the result would be a subjective evaluation of the performance. This study proposes instead a preliminary analysis of the performance of motion cueing algorithms with objective evaluation by means of a human motion perception model.@en

The measurement and estimation of wheel loads is an interesting and complex topic relevant for vehicle dynamics control. Accurate wheel load information allows for more straight-forward, more robust and more efficient control. In this paper a novel model based wheel load reconstruction approach is presented. An Unscented Kalman Filter is used to reconstruct the unknown wheel loads by analysis of the deformation of the bearing outer-ring. The performance of the approach is demonstrated by field tests using an instrumented passenger car. Results show that the proposed approach is well able to reconstruct both tilting and self-aligning moments as well as lateral and vertical wheel forces during various steering maneuvers.@en

This work discusses a road-tyre friction estimator considering combined tyre slip. The friction estimator design is motivated by its importance in vehicle dynamics control as accurate friction estimation can improve performance and safety. The estimator uses tyre force measurements from Load Sensing Bearing (LSB) technology and does not rely on parameterized tyre model. The tyre force measurements benefit the estimator mainly because of the uncertainties and nonlinearities of the tyre force characteristics. The proposed estimator uses tyre slip and tyre force representations where the longitudinal and lateral tyre slips and forces are combined into a single tyre slip and tyre force values. This representation makes the method effective during pure longitudinal dynamics, pure lateral dynamics and for combined slip. In addition, individual tyre-road friction estimation is possible with the proposed estimator and a computationally inexpensive algorithm, suitable for real-time implementation, is used to estimate the friction. The estimator is studied in simulation during pure braking, pure cornering and for combined slip. Further, the estimator is simulated in closed loop with a yaw rate controller to study whether the estimator improves vehicle safety. Finally the estimator is validated using test data from several maneuvers performed on a test vehicle instrumented with LSB technology.

@en

This paper investigates the potential of load based vehicle sideslip estimation. Different techniques to measure tyre forces have been presented over the years; so far no technique has made it to the market. This paper considers a new technology based on load sensing bearings, which provides tyre force measurements. Based on the features of the sensor, a vehicle sideslip angle estimator is designed, analyzed and tested. The paper shows that direct tyre force sensing has mainly two advantages over traditional model-based estimators: primarily, it avoids the use of tyre models, which are heavily affected by uncertainties and modeling errors and secondarily, providing measurements on the road plane, it is less prone to errors introduced by roll and pitch dynamics. Extensive simulation tests along with a detailed analysis of experimental tests performed on an instrumented vehicle prove that the load based estimation outperforms the kinematic model-based benchmark yielding a root mean square error of 0.15°.

@en

Active vehicle safety and driving assistance systems can be made more efficient, more robust and less complex if wheel load information would be available. Although this information could be determined via numerous different methods, due to various reasons, no commercially feasible approach has yet been introduced. In this paper the approach of bearing load estimation is topic of interest. Using the bearing for load measurement has considerable advantages making it commercially attractive as: i) it can be performed on a non-rotating part, ii) all wheel loads can be measured and iii) usually the bearing serves the entire lifetime of the vehicle. This paper proposes a novel approach for the determination of wheel loading. This new approach, based on the strain variance on the surface of the bearing outer ring, is tested on a dedicated bearing test setup. The experimental results show that the approach allows for the determination of all three force vectors and two moments in different load conditions with sufficient accuracy and bandwidth suitable for the application in vehicle dynamics control. @en

Research objective: Anti-lock braking algorithms use either/both wheel deceleration and wheel slip to obtain a stable limit cycle around the friction peak to guarantee vehicle steerability and to minimize braking distance. However, both control variables pose several well-known issues regarding ABS control. The usage of wheel loads, for instance estimated based on bearing deformation, could provide a solution to these control variable related difficulties. In this paper, a wheel load based method to control wheel slip is presented and implemented in a novel Anti-lock Braking algorithm. Due to the fundamentally different approach to tackle the issue, numerous well known pitfalls of traditional Anti-lock Braking Systems can be avoided. Methodology: A mathematical derivation of the quarter car model provides the conditions in which wheel load measurement allows for determination of the derivative of wheel slip. Based on this theory, a novel ABS algorithm is proposed. It consists of two operational phases to control the wheel slip derivative and a phase switching mechanism, all based solely on wheel loads. Furthermore a methodology of wheel load estimation based on bearing deformation measurement is proposed. Finally, an experimental on-road investigation of the load estimation and proposed algorithm is carried out using an instrumented test vehicle. Results: An on-road investigation with a test vehicle demonstrates the accuracy of wheel load estimation based on bearing deformation. The estimated loads are used in a novel ABS algorithm to demonstrate the feasibility and advantages of load based ABS control. Limitations of this study: Only straight-line braking is considered as the method of load estimation is currently unable to provide the required bandwidth on estimation of loads when steering. What does the paper offer that is new in the field: Current research in the field of ABS algorithms is primarily focused on wheel slip and/or wheel deceleration control. The presented study investigates a fundamentally different approach by the use of a novel sensor. Conclusion: Based on a mathematical derivation a novel load-based ABS algorithm is proposed. Furthermore a methodology of load sensing by the use of instrumented bearings is presented. The performance of both load sensing and the Anti-lock braking algorithm has been checked via experimental testing using an instrumented test vehicle.@en

This paper supplies a roadmap on how a researcher can effectively perform real vehicular experiments oriented to high-speed driving research. It provides detailed guidelines for constructing versatile low-cost instrumentation suitable to be fitted on race cars. The custom-built equipment, consisting of wheel-speed sensors, steering angle-torque sensors, electronic boards, etc., is thoroughly described. Furthermore, this paper depicts the required processing from raw measurements to user-friendly data suitable for driver behavior studies. As an illustration, a case study on driving behavior analysis is presented, during the execution of high-speed circular maneuvers. The recorded data showed markedly different driving behaviors between expert and novice drivers. The mechanical designs and the open-source-based software are freely available online.

@en