Currently, the objective evaluation of the DCT vehicle drivability requires the accurate identification of the driver’s intention and vehicle state as well as the selection of the targeted evaluation indicators. The existing identification methods usually cannot divide the driver’s intentions in detail and make full use of the characteristics of time-series signals. Simultaneously, external kinematic sensors are more commonly used than the sensors of vehicle powertrain, which impacts the recognition effect. This paper proposes a new method for identifying the DCT vehicle driver’s starting intentions based on an LSTM neural network and multi-sensor data fusion. The DCT vehicle driver’s starting intentions are subdivided and defined based on human–vehicle interaction analysis and K-means clustering. The input of the model consists of 11-dimensional variables that include motion parameters of the vehicle collected by the external sensors and the powertrain parameters collected by onboard sensors. The method proposed in this paper first establishes a recognition window, which is utilized to extract the starting process samples from the DCT vehicle driving data. Second, the 11 variables of each sample are used as one set of multi-dimensional time-series signals, which are preprocessed through wavelet denoising. Finally, the LSTM network is used to identify the samples. The identification results indicate that the highest recognition accuracy of the proposed algorithm is 94.27%, which is approximately 5% higher than conventional methods, such as fully connected neural networks and support vector machines. Furthermore, the model with 11 input variables outperforms the model with fewer input variables. The effectiveness and superiority of the identification model have been demonstrated.

In the road test of vehicle performance evaluation, real-time and accurate estimation of road slope is essential for objective evaluation. Using slope meter directly can bring many problems such as large randomness and errors in road test. Using complex road slope estimation algorithm often brings redundant sensors and reduces detection efficiency. Aiming at the above problems, this paper proposes a road slope estimation model based on IMU error calibration and multi-sensor signal fusion. First, the vehicle-road dynamics and kinematics models are established. Then, the error sources of IMU are analyzed, and the calibration and compensation methods are proposed. The acceleration signal of IMU is compensated by inertia through the vehicle velocity signal obtained by multi-sensors, and the projection of gravity acceleration vector in the vehicle coordinate is decoupled. Finally, the model fuses the decoupled result with IMU angular velocity value through Kalman filter algorithm, and outputs the estimated slope of the road. The road test results show that the model can effectively compensate the IMU installation error and accurately estimate the road slope. And the slope estimation error is less than 0.5%, which can meet the needs of the road test of vehicle performance evaluation.

Aiming at the problem that the base coordinate system of industrial robot is not unified with the center coordinate system of tool, and it is impossible to input the action position in the robot teaching device, this paper adopts ROMER HEXAGON METROLOGY 7530SE three-coordinate measuring instrument as the robot measuring equipment and proposes a multi-objective normalization algorithm. In this paper, the reliability of the multi-objective normalization algorithm is verified by the experiment, and the measurement accuracy difference of the multi-objective normalization algorithm under different reference systems is studied. The results have shown that the accuracy of the system reaches 0.8 mm and the root mean square (RMS) value is maintained at 0.3-0.7 mm when the engine is taken as the reference coordinate system within the robot's 0-50mm motion stroke. When the body-in-white is used as the reference coordinate system, the accuracy of the system can reach 2.2 mm, the RMS value can be maintained at 1.0-1.8mm. The reasons for good accuracy and stability of the measurement system established with the engine as the reference coordinate are analyzed. The multi-objective normalization algorithm proposed in this paper has high engineering universality, and the accuracy analysis of the algorithm under different reference system has guiding significance for the selection of reference coordinate system of measurement system.