Search results

Feature selection is crucial for machine learning to recognize lane-change (LC) maneuver as there exist a large number of feature candidates. Blindly using feature could take up large storage and excessive computation time, while insufficient feature selection would cause poor performance. Selecting high contributive features to classify LC and lane-keep behavior is effective for maneuver recognition. This paper aims to propose a feature selection method from a statistical view based on an analysis from naturalistic driving data.

In total, 1,375 LC cases are analyzed. To comprehensively select features, the authors extract the feature candidates from both time and frequency domains with various LC scenarios segmented by an occupancy schedule grid. Then the effect size (Cohen’s d) and p-value of every feature are computed to assess their contribution for each scenario.

It has been found that the common lateral features, e.g. yaw rate, lateral acceleration and time-to-lane crossing, are not strong features for recognition of LC maneuver as empirical knowledge. Finally, cross-validation tests are conducted to evaluate model performance using metrics of receiver operating characteristic. Experimental results show that the selected features can achieve better recognition performance than using all the features without purification.

In this paper, the authors investigate the contributions of each feature from the perspective of statistics based on big naturalistic driving data. The aim is to comprehensively figure out different types of features in LC maneuvers and select the most contributive features over various LC scenarios.

This paper is concerned with the repetitive trajectory tracking control for space manipulators under model uncertainties and vibration disturbances.

The model uncertainties and link vibration of manipulators will degrade the tracking performance of space manipulators; in this paper, a new hybrid control scheme that consists of a composite hierarchical anti-disturbance controller and an iterative learning controller is developed to solve this problem.

The composite hierarchical controller can effectively attenuate model uncertainties and reject vibration disturbances, whereas the iterative learning controller is able to improve the tracking accuracy for repetitive reference trajectory.

The proposed scheme compensates for the shortcomings of iterative learning control which can only deal with repetitive disturbances, ensuring the accuracy and repeatability of space manipulators under model uncertainties and random disturbances.

Construction schedule delays and quality problems caused by construction errors are common in the field of prefabricated buildings. The effective monitoring of the construction project process is one of the key factors for the success of a project. How to effectively monitor the construction process of prefabricated building construction projects is an urgent problem to be solved. Aiming at the problems existing in the monitoring of the construction process of prefabricated buildings, this paper proposes a monitoring method based on the feature extraction of point cloud model.

This paper uses Trimble X7 3D laser scanner to complete field data collection experiments. The point cloud data are preprocessed, and the prefabricated component segmentation and geometric feature measurement are completed based on the PCL platform. Aiming at the problem of noisy points and large amount of data in the original point cloud data, the preprocessing is completed through the steps of constructing topological relations, thinning, and denoising. According to the spatial position relationship and geometric characteristics of prefabricated frame structure, the segmentation algorithm flow is designed in this paper. By processing the point cloud data of single column and beam members, the quality of precast column and beam members is measured. The as-built model and as-designed model are compared to realize the visual monitoring of construction progress.

The experimental results show that the dimensional measurement accuracy of beam and column proposed in this paper is more than 95%. This method can effectively detect the quality of prefabricated components. In the aspect of progress monitoring, the visualization of real-time progress monitoring is realized.

This paper proposed a new monitoring method based on feature extraction of the point cloud model, combined with three-dimensional laser scanning technology. This method allows for accurate monitoring of the construction process, rapid detection of construction information, and timely detection of construction quality errors and progress delays. The treatment process based on point cloud data has strong applicability, and the real-time point cloud data transfer treatment can guarantee the timeliness of monitoring.

The purpose of this paper is to estimate the deformation of soft manipulators caused by obstacles accurately and the contact force and workspace can be also predicted.

The continuum deformation of the backbone of the soft manipulator under contact is regarded as two constant curvature arcs and the curvatures are different according to the fluid pressure and obstacle location based on piecewise constant curvature framework. Then, this study introduces introduce the moment balance and energy conservation equation to describe the static relationship between driving moment, elastic moment and contact moment. Finally, simulation and experiments are carried out to verify the accuracy of the proposed model.

For rigid manipulators, environmental contact except for the manipulated object was usually considered as a “collision” which should be avoided. For soft manipulators, an environment is an important tool for achieving manipulation goals and it might even be considered to be a part of the soft manipulator’s end-effector in some specified situations.

There are also some limitations to the presented study. Although this paper has made progress in the static modeling under environmental contact, some practical factors still limit the further application of the model, such as the detection accuracy of the environment location and the deformation of the contact surface.

Based on the proposed kinematic model, the bending deformation with environmental contact is discussed in simulations and has been experimentally verified. The comparison results show the correctness and accuracy of the presented SCC model, which can be applied to predict the slender deformation under environmental contact without knowing the contact force.