A novel multi-sensors fusion framework based on Kalman Filter and neural network for AFS application

Abstract

An adaptive front light system (AFS) is put forward by the Society of Automotive Engineers and Economic Commission for Europe as a means of enhancing vehicular lighting. Traditionally, AFS can be divided into three parts: (1) a leveling subsystem to make lighting parallel to the road surface; (2) a swiveling subsystem to change light distribution along with the angle of the steering wheel; (3) a dimming subsystem to reduce or intensify the lighting. In this paper, a new hybrid multi-sensor fusion framework combining Kalman Filter with neural network is proposed to adjust two stepper motors controlling the vehicles headlights pitch and yaw. Kalman Filter as the frontend is used to deal with redundant sensor signals that are collected from sensors in the different places. Fuzzy Neutral Network as the backend is used to generate adjustment of leveling and swiveling angle through the integration of different type signals. An adaptive parameter adjustment is accomplished by the proposed fusion framework with the varying filter coefficients. The simulation and experiment of leveling angle are conducted using the predefined experimental data. The evaluation results of leveling angle prove that the proposed algorithm can effectively filter out high-frequency perturbations and provide reliable outputs for stepper motor. The same results can be obtained for a swiveling subsystem. Consequently, the hybrid fusion framework is a feasible approach for AFS design to accomplish data processing and nonlinear mapping.

A Neuron-Based Kalman Filter with Nonlinear Autoregressive Model

“…Hu et al [49] estimated the target location with Kalman filter and the estimation was imported into BPNN to classify the targets. Liu et al [50] utilized Kalman filter and fuzzy neural network (FNN) in a multi-source data fusion framework of an adaptive control system, in which data was processed firstly with Kalman filter, and the filtered results were set as the input of FNN. Others [26,27,51] used a Kalman filter and ANN in reverse order, in which ANN is constructed before the Kalman filter.…”

Advanced Estimation Techniques for Vehicle System Dynamic State: A Survey

“…In [137], the adaptive neuro-fuzzy inference system (ANFIS) with input delay technique was developed to estimate vehicle velocity and position through the fusion of datasets from the GPS and inertial navigation system (INS); experimental results have demonstrated that ANFIS can provide improved estimation accuracy when compared with the EKF method. The NNs have also been employed to estimate vehicle states by fusing multi-sensors [138,139,140,159]. The integration of GPS/INS through NNs considered in [139] was done to process the GPS signal in case of INS signal loss so that it can obtain accurate position and data, whereas the neural network-based MMF in [140] was adopted in order to obtain accurate and reliable position estimation of autonomous vehicles by combining GPS and on-board sensors.…”

A self-adaptive distributed decision support model for Internet of Things applications

“…In this kind of approach, decision making is carried out by one or more central units (López-Ortega and Rosales, 2011; Qin and Liu, 2015). This is usually used in control-related applications, such as an air temperature controller (Liu et al, 2015; Zhu et al, 2015). The related research normally focuses on the fusion and coordination of heterogeneous information (Bui and Lee, 1999; Sokolova and Fernández-Caballero, 2009).…”