Zengcai Wang

Shearer positioning technology can effectively improve the automation level of fully mechanized mining face. The Strapdown Inertial Navigation System (SINS) and odometer integrated navigation system assisted by nonholonomic constraint (NHC) and closed path optimal estimation model (CPOEM) has been widely studied in shearer positioning. Due to the difficulty in accurately measuring and controlling the distance and direction of the hydraulic support, abnormal errors often occur between the predicted position calculated by CPOEM and the actual position of shearer, and measurement noise cannot be accurately determined, thereby seriously reducing the positioning accuracy of shearer. To address this issue, an adaptive and robust shearer navigation system based on the interacting multiple model (IMM) algorithm is proposed in this paper. Firstly, maximum correntropy criterion Kalman filter (MCCKF) is used to replace KF to reject outliers in predicted positions, and adaptive kernel bandwidth is introduced to MCCKF to further enhance robustness; Secondly, Sage-Husa Adaptive Kalman Filter (SHAKF) is used to estimate the measurement noise in the predicted position in real-time; Finally, MCCKF and SHAKF are used as two sub filters of IMM working in parallel, and the state estimation of each sub filter is fused based on the quality of innovation. Through simulation and experiment, it has been proven that the navigation scheme proposed in this paper can integrate the advantages of MCCKF and SHAKF, avoiding the interference of outliers and estimating measurement noise in real-time, effectively improving the positioning accuracy of shearer.

The integrated navigation system of strapdown inertial navigation system (SINS) and odometer (OD) is a commonly used vehicle positioning system. The installation errors caused by the difference between the installation positions of SINS and OD mainly include the installation deflection angle and lever arm, which will reduce the positioning accuracy. The traditional calibration method is to expand the installation deflection angle error and lever arm error into state variables, the velocity difference under nonholonomic constraint (NHC) is taken as the measurement to establish the Kalman filter model for online calibration, but this method seriously depends on the sensor accuracy. In order to calibrate the installation errors of low-precision SINS, this paper improves NHC, adds the difference between the centripetal acceleration measured by odometer and calculated by SINS as measurement on the basis of the original measurement, and introduces a low-pass filter to reduce the interference of vehicle vibration on the accelerometer.The experimental results show that the proposed algorithm has better convergence of the installation deflection angle, and the error between the estimation results of the forward lever arm and the actual measurement results is less than 0.1m; After compensating the installation errors of the improved NHC, the maximum positioning errors in the east and north directions are reduced by 63.19% and 43.51% respectively, the height error is also greatly reduced, and the positioning accuracy of the integrated navigation system is significantly improved.

Clutch control has a great effect on the starting quality and shifting quality of heavy-duty vehicles with automated mechanical transmission (AMT). The motion characteristics of a clutch actuator for heavy-duty vehicles with AMT are studied in this paper to investigate the clutch control strategy further. The modeling principle of the automatic clutch system is analyzed, and a simulation analysis is given to prove its validity and rationality. Normalized velocity and velocity modulation percentage are proposed as evaluation parameters for the clutch actuator driven by pulse width modulation (PWM) signals. Based on an AMT test bench, the actuator motion characteristics are analyzed. Experimental results show that the range of normalized velocity and velocity modulation percentage are obtained for the clutch engagement and disengagement processes. By analyzing the experimental data, the engaging velocity and disengaging velocity of the actuator are estimated using the solenoid valves in combination. The research results provide a fundamental basis for precise controlling of the clutch and improving the smoothness of heave-duty vehicles.

In this paper, a novel adaptive cruise control (ACC) algorithm based on model predictive control (MPC) and active disturbance rejection control (ADRC) is proposed. This paper uses an MPC algorithm for the upper controller of the ACC system. Through comprehensive considerations, the upper controller will output desired acceleration to the lower controller. In addition, to increase the accuracy of the predictive model in the MPC controller and to address fluctuations in the vehicle’s acceleration, an MPC aided by predictive estimation of acceleration is proposed. Due to the uncertainties of vehicle parameters and the road environment, it is difficult to establish an accurate vehicle dynamic model for the lower-level controller to control the throttle and brake actuators. Therefore, feed-forward control based on a vehicle dynamic model (VDM) and compensatory control based on ADRC is used to enhance the control precision and to suppress the influence of internal or external disturbance. Finally, the proposed optimal design of the ACC system was validated in road tests. The results show that ACC with APE can accurately control the tracking of the host vehicle with less acceleration fluctuation than that of the traditional ACC controller. Moreover, when the mass of the vehicle and the slope of the road is changed, the ACC–APE–ADRC controller is still able to control the vehicle to quickly and accurately track the desired acceleration.

The key technology to realize intelligent unmanned coal mining is the strapdown inertial navigation system (SINS); however, the gradual increase of cumulative error during the working process needs to be solved. On the basis of an SINS/odometer (OD)-integrated navigation system, this paper adds magnetometer (MAG)-aided positioning and proposes an SINS/OD/MAG-integrated shearer navigation system. The velocity observation equation is obtained from the speed constraints during shearer movement, and the yaw angle observation equation is obtained from the magnetometer output. The position information of the SINS output is calibrated using these two observations. In order to improve the fault tolerance of the integrated navigation system, an adaptive federated Kalman filter is established to complete the data fusion of the SINS. Experimental results show that the positioning accuracy of the SINS/OD/MAG-integrated navigation system is 75.64% and 74.01% higher in the east and north directions, respectively, than the SINS/OD-integrated navigation system.

The effective detection of driver drowsiness is an important measure to prevent traffic accidents. Most existing drowsiness detection methods only use a single facial feature to identify fatigue status, ignoring the complex correlation between fatigue features and the time information of fatigue features, and this reduces the recognition accuracy. To solve these problems, we propose a driver sleepiness estimation model based on factorized bilinear feature fusion and a long- short-term recurrent convolutional network to detect driver drowsiness efficiently and accurately. The proposed framework includes three models: fatigue feature extraction, fatigue feature fusion, and driver drowsiness detection. First, we used a convolutional neural network (CNN) to effectively extract the deep representation of eye and mouth-related fatigue features from the face area detected in each video frame. Then, based on the factorized bilinear feature fusion model, we performed a nonlinear fusion of the deep feature representations of the eyes and mouth. Finally, we input a series of fused frame-level features into a long-short-term memory (LSTM) unit to obtain the time information of the features and used the softmax classifier to detect sleepiness. The proposed framework was evaluated with the National Tsing Hua University drowsy driver detection (NTHU-DDD) video dataset. The experimental results showed that this method had better stability and robustness compared with other methods.

The technology of coal-rock interface recognition is the core of realizing the automatic heightening technology of shearer’s rocker. Only by accurately and quickly identifying the interface of coal and rock can we realize the fully automatic control of shearer. As the only one used in the actual detection of coal mining machine drum cutting coal seam after the thickness of the remaining coal seam detection method, natural γ-ray has a very practical advantage. Based on the relationship between the attenuation of the natural γ-ray passing through the coal seam and the thickness of the coal seam, the mathematical model of the attenuation of the natural γ-ray penetrating coal seam is established. By comparing the attenuation intensity of γ-ray with or without brackets, it is verified that the hydraulic girders will absorb some natural γ-rays. Finally, this paper uses the ground simulation experiment and the field experiment to verify the correctness of the mathematical model and finally develop the natural γ-ray seam thickness sensor. The sensor has the function of indicating the thickness of the coal seam, measuring the natural γ-ray intensity, and storing and processing the data.

When a low-cost micro-electro-mechanical system inertial measurement unit is used for a vehicle navigation system, errors will quickly accumulate because of the large micro-electro-mechanical system sensor measurement noise. To solve this problem, an automotive sensor–aided low-cost inertial navigation system is proposed in this article. The error-state model of the strapdown inertial navigation system has been derived, and the measurements from the wheel speed sensor and steer angle sensor are used as the new observation vector. Then, the micro-electro-mechanical system inertial measurement unit/wheel speed sensor/steer angle sensor–integrated system is established based on adaptive Kalman filtering. The experimental results show that the positioning error of micro-electro-mechanical system inertial measurement unit/wheel speed sensor/steer angle sensor is 94.67%, 98.88%, and 97.88% less than the values using pure strapdown inertial navigation system in the east, north, and down directions, respectively. The yaw angle error is reduced to less than 1°, and the vehicle velocity estimation of micro-electro-mechanical system inertial measurement unit/wheel speed sensor/steer angle sensor–integrated navigation system is closer to the reference value. These results show the precision of the integrated navigation solution.

This paper presents a lane departure detection approach that utilizes a stacked sparse autoencoder (SSAE) for vehicles driving on motorways or similar roads. Image preprocessing techniques are successfully executed in the initialization procedure to obtain robust region-of-interest extraction parts. Lane detection operations based on Hough transform with a polar angle constraint and a matching algorithm are then implemented for two-lane boundary extraction. The slopes and intercepts of lines are obtained by converting the two lanes from polar to Cartesian space. Lateral offsets are also computed as an important step of feature extraction in the image pixel coordinate without any intrinsic or extrinsic camera parameter. Subsequently, a softmax classifier is designed with the proposed SSAE. The slopes and intercepts of lines and lateral offsets are the feature inputs. A greedy, layer-wise method is employed based on the inputs to pretrain the weights of the entire deep network. Fine-tuning is conducted to determine the global optimal parameters by simultaneously altering all layer parameters. The outputs are three detection labels. Experimental results indicate that the proposed approach can detect lane departure robustly with a high detection rate. The efficiency of the proposed method is demonstrated on several real images.