An Extended Kalman Filter and Back Propagation Neural Network Algorithm Positioning Method Based on Anti-lock Brake Sensor and Global Navigation Satellite System Information

Abstract

Telematics box (T-Box) chip-level Global Navigation Satellite System (GNSS) receiver modules usually suffer from GNSS information failure or noise in urban environments. In order to resolve this issue, this paper presents a real-time positioning method for Extended Kalman Filter (EKF) and Back Propagation Neural Network (BPNN) algorithms based on Antilock Brake System (ABS) sensor and GNSS information. Experiments were performed using an assembly in the vehicle with a T-Box. The T-Box firstly use automotive kinematical Pre-EKF to fuse the four wheel speed, yaw rate and steering wheel angle data from the ABS sensor to obtain a more accurate vehicle speed and heading angle velocity. In order to reduce the noise of the GNSS information, After-EKF fusion vehicle speed, heading angle velocity and GNSS data were used and low-noise positioning data were obtained. The heading angle speed error is extracted as target and part of low-noise positioning data were used as input for training a BPNN model. When the positioning is invalid, the well-trained BPNN corrected heading angle velocity output and vehicle speed add the synthesized relative displacement to the previous absolute position to realize a new position. With the data of high-precision real-time kinematic differential positioning equipment as the reference, the use of the dual EKF can reduce the noise range of GNSS information and concentrate good-positioning signals of the road within 5 m (i.e. the positioning status is valid). When the GNSS information was shielded (making the positioning status invalid), and the previous data was regarded as a training sample, it is found that the vehicle achieved 15 minutes position without GNSS information on the recycling line. The results indicated this new position method can reduce the vehicle positioning noise when GNSS information is valid and determine the position during long periods of invalid GNSS information.

Research on a Wi-Fi RSSI Calibration Algorithm Based on WOA-BPNN for Indoor Positioning

“…Selami et al [13] employed a BPNN model to reduce the effect of the nonlinearities presented in laser triangulation displacement sensors. Hu et al [14] developed a BPNN model to correct heading angle velocity output and vehicle speed and to add the synthesized relative displacement to the previous absolute position to realize a new vehicle position. Moreover, based on BPNN model, Xing et al [15] proposed a 60 GHz impulse radio positioning algorithm, which has obtained a good positioning result.…”

Asynchronous Sensor Fusion of GPS, IMU and CAN-Based Odometry for Heavy-Duty Vehicles

“…A well-known method to fuse data measurements from different sensors is the Kalman filter (KF). In tracking and egomotion estimation applications [10], [11], [12], this method updates the estimation of a state, usually containing position and orientation of an object, based on sensor measurements. Typically used variables in the state of these filters in vehicles are position, linear and angular velocities and accelerations [13], but there are situations where biases of sensors are also estimated to get a better overall estimation [14], [15], [16].…”

Driving Performance Assessment Based on Accurate Position Tracking

“…BeiDou Navigation Satellite System (BDS) [16,17] has been in rapid expansion since its first launch in 2000. As an alternative of GPS in the Asia-Pacific region, BDS has been widely used in transportation [18], agriculture [19], disaster prevention and mitigation [20], and many more. e tracking module includes a fixed base station and a mobile base station, where the fixed base station receives BeiDou satellite ephemeris data, and the mobile base station, which is mounted on the vehicle, receives differential data from the fixed base station.…”