Top Read Articles

Published in last 1 year |

In last 2 years |

In last 3 years |

All

Please wait a minute...


For Selected: Download Citations EndNote Ris BibTeX Toggle Thumbnails

Select

A Robot Grasping Policy Based on Viewpoint Selection Experience Enhancement Algorithm WANG Gao, CHEN Xiaohong, LIU Ning, et al Journal of South China University of Technology(Natural Science Edition)    2022, 50 (9): 126-137.   DOI: 10.12141/j.issn.1000-565X.210769 Abstract （10017）   HTML （20）    PDF（pc） （3882KB）（1098）       Knowledge map    Save To solve the problem of the low success rate of robot vision grasping using fixed environment camera in the scene of cluttered and stacked objects, an eye-hand follow-up camera viewpoint selection policy based on deep reinforcement learning is proposed to improve the accuracy and speed of vision-based grasping. Firstly, a Markov decision process model is constructed for robot active vision-based grasping task, then the problem of viewpoint selection is transformed into a problem of solving the viewpoint value function. A deconvolution network with encoder-decoder structure is used to approximate the viewpoint action value function, and the reinforcement learning is carried out based on the deep Q-network framework. Then, to resolve the problem of sparse reward existing in reinforcement learning, a novel viewpoint experience enhancement algorithm is proposed. The different enhancement methods between the successful and failed grasping process are designed respectively. And the reward region can be expanded from a single point to a circular region for improving the convergence speed of the approximation network. The preliminary experiment is deployed on the simulation platform, and the robot model and the grasping environment are simultaneously built in the simulation platform to implement the offline reinforcement learning. In the process, the proposed viewpoint experience enhancement algorithm can effectively improve the sample utilization rate and speed up the convergence of training. Based on the proposed viewpoint experience enhancement algorithm, the viewpoint action value function approximation network can converge within 2 h. To obtain the results from the verification with application, the proposed viewpoint selection policy is applied to the real-world scenes with robot for grasping experiments. The result shows that the viewpoint optimization based on this policy can effectively promote the accuracy and speed of robot grasping. Compared with the general grasping methods, the proposed viewpoint selection policy needs only one viewpoint selection in real-world robot grasping to find the focus region with high grasping success rate. And the method can also promote the processing efficiency of the best viewpoint selection. The grasping success rate in cluttered scenes is increased by 22.8% against the single-view method, and the mean picks per hour can reach 294 units. As whole, it shows that the proposed policy has the capacity of industrial application. Table and Figures | Reference | Related Articles | Metrics | Comments（0）

Select

Research on Autonomous Grasping of a Humanoid Robot Based on Vision ZHANG Xuchong, YANG Jun Journal of South China University of Technology(Natural Science Edition)    2024, 52 (7): 53-61.   DOI: 10.12141/j.issn.1000-565X.230528 Abstract （8009）   HTML （20）    PDF（pc） （3170KB）（272）       Knowledge map    Save With the development of robotics technology, humanoid robots have shown application potential and value in multiple fields. Research on autonomous grasping of humanoid robots based on machine vision aims to improve their grasping adaptability and humanoid actions in natural environments. In terms of machine vision, the Realsense-D435 depth camera was adopted, and the YOLO (You Only Look Once) object detection model was used to achieve target object recognition, spatial positioning, depth map cropping, and target point cloud generation. The object’s posture was obtained based on the registration algorithm (ICP) between the target point cloud and the standard point cloud. The robot head was modeled using the D-H method, and the position and posture of the object were converted from the camera coordinate system to the robot coordinate system. In terms of motion planning, according to the grasping law of the human arm, the grasping process was divided into 9 basic actions: initial position, moving to the pre-grasping position, grasping the object, lifting the object, moving the object, moving to placement position placing the object, retreating position, and returning to the initial position. Corresponding grasping postures were determined for different objects to improve the success rate of grasping. Based on the grasping and placing points obtained visually, the remaining key points were calculated independently, and the spatial arc was used as the grasping trajectory. Through Matlab simulation, the rationality of the end movement trajectory and joint trajectory of the robotic arm during the grasping process was verified. Finally, an object grasping experiment was conducted, and the results showed that the humanoid robot can quickly and accurately recognize and locate different objects in the natural environment, and can successfully grasp and transport them with a success rate of over 80%. And it takes into account the imitation of human nature of the action, verifying the effectiveness of the proposed solution. This study can promote the application and popularization of humanoid robots in human daily life. Table and Figures | Reference | Related Articles | Metrics | Comments（0）

Select

Study on Measuring Method of Vehicle Carbon Emission in Expressway Network LIN Xukun, ZHANG Yang, LUO Zhiqing, et al Journal of South China University of Technology(Natural Science Edition)    2022, 50 (9): 22-28.   DOI: 10.12141/j.issn.1000-565X.220019 Abstract （5481）   HTML （27）    PDF（pc） （1685KB）（1140）       Knowledge map    Save Under the background of dual-carbon policy, it is imperative to maintain ecological balance and reduce carbon emissions. Nowadays, it is very important to accurately measure the carbon emission index of vehicles on a wide range of road networks. Therefore, this study proposed a measuring method of vehicle carbon emission in expressway network based on multi-source data fusion. Firstly, a basic data cleaning method for carbon emission statistics was proposed to clean the basic data required for subsequent carbon emission calculation. Secondly, the highway carbon emission calculation model was established, and then the related calculation process was designed. Finally, taking the whole highway network of Guangdong province as an example, this paper calculated the vehicle carbon emission from September 2020 to June 2021, and compared the calculation results with China’s carbon accounting database.Through this method, the proposed method was proved to be scientific and reliable. The research shows that the average carbon emission of mini buses in Guangdong province is small, but the total carbon emission accounts for the largest proportion of all types of vehicles, up to 52.1%; the total carbon emission of gasoline vehicles accounts for 49.8%, which is higher than that of diesel vehicles (45.4%) and of other energy vehicles (4.8%). Vigorously promoting new energy vehicles can effectively reduce the carbon emission of expressways. In addition, the study finds that there are significant differences in the travel patterns of different vehicles under COVID-19, but the overall impact on the transportation economy is limited. Table and Figures | Reference | Related Articles | Metrics | Comments（0）

Select

Research Progress on Key Technologies in the Cooperative Vehicle Infrastructure System LIN Hongyi, LIU Yang, LI Shen, et al Journal of South China University of Technology(Natural Science Edition)    2023, 51 (10): 46-67.   DOI: 10.12141/j.issn.1000-565X.230200 Abstract （4938）   HTML （30）    PDF（pc） （1996KB）（427）       Knowledge map    Save With the steady growth of urban car ownership, the issue of traffic congestion is becoming increasingly prominent, bringing great pressure to urban development. To respond effectively to this challenge, it is critical to develop methods that can improve transport efficiency and reduce energy consumption. In current context, the Cooperative Vehicle Infrastructure System (CVIS), an ideal solution for realizing green and intelligent transportation systems, has become an important direction in both transportation research and practice. By integrating and optimizing various traffic resources, CVIS not only enhances traffic efficiency and reduces energy consumption but also provides key technical support for achieving “dual carbon” goals. This paper thoroughly analyzed the fundamental concepts, research methodologies and application scenarios of CVIS, and delved into its four core technological modules: fusion perception, driving cognition, autonomous decision-making, and cooperative control. The paper reviewed and summarized research achievements within these modules, ranging from traditional methods to the latest in deep reinforcement learning techniques. It also explored the potential applications of these technologies and methods for enhancing traffic efficiency, reducing energy consumption, and improving road safety. Finally, the paper scrutinized numerous challenges that CVIS may encounter in practical applications, including the security of information transmission, system stability, and environmental complexity. To overcome these challenges, the paper looked forward to the future development in four areas: developing datasets that integrate vehicle-side and roadside information, enhancing the fusion accuracy of multi-source perception information, improving the real-time performance and safety of CVIS, and optimizing multi-vehicle cooperative decision-making control methods under complex conditions. As a result, this paper not only has important reference value for the advancement of CVIS technology, but also provides important guidance for the future planning and construction of urban transportation systems. Table and Figures | Reference | Related Articles | Metrics | Comments（0）

Select

Optimization Model of Bus Priority Control Considering Carbon Emissions with Stochastic Characteristics HU Xinghua, CHEN Xinghui, WANG Ran, et al Journal of South China University of Technology(Natural Science Edition)    2023, 51 (10): 160-170.   DOI: 10.12141/j.issn.1000-565X.230178 Abstract （4577）   HTML （3）    PDF（pc） （2474KB）（104）       Knowledge map    Save In the context of the construction of a country with a strong transportation network, vigorously developing urban public transportation and promoting sustainable urban development has become an inevitable requirement for urban transportation development. Transit signal priority control, as an active priority strategy, can effectively reduce the carbon emissions and delays generated by buses at signal intersections, and improve the quality of bus service. A bus speed probability density function was introduced to study the effect of bus priority control strategy on traffic carbon emission, based on the speed stochastic characteristics of intersection. The effect of main parameters such as delay, stopping times, and speed on traffic carbon emission was analyzed. A bi-level optimization model of single-intersection bus priority control was established using the combination strategy of speed guidance and green extension. The model took the optimal carbon emission reduction of buses and cars with different fuel types in the upstream section of the intersection and the intersection control area as the upper-level objective, the optimal total people delay reduction as the lower-level objective, and the guidance speed as well as the compressed green time of the non-bus-priority phases as the decision variables. The Gauss-Seidel iterative algorithm was used to solve the model. Finally, the established model was applied to the calculation cases for analysis, and the results indicated that under the guidance acceleration and green extension strategy, the overall carbon emission and total passenger delay reduction of the intersection could reach 25.63% and 36.27%, respectively. The model effectively reduced carbon emissions and total passenger delays in the upstream sections of the intersection and the intersection control area, and optimized the overall traffic benefit of the intersection while promoting sustainable development. Table and Figures | Reference | Related Articles | Metrics | Comments（0）

Select

Study on Accident Risk Based on Driving Behavior and Traffic Operating Status GUO Miao, ZHAO Xiaohua, YAO Ying, et al Journal of South China University of Technology(Natural Science Edition)    2022, 50 (9): 29-38.   DOI: 10.12141/j.issn.1000-565X.210629 Abstract （4303）   HTML （29）    PDF（pc） （2302KB）（1149）       Knowledge map    Save Accurate identification of traffic accident risk and timely mastering the change of traffic crash risk are of great significance for proactive prevention and reduction of traffic accident. Most of the existing traffic crash identification studies are based on real-time and dynamic parameters such as traffic flow and traffic conflict. The application of risky driving behavior in traffic accident risk detection is limited by the constraints of previous data acquisition technologies. To more accurately identify the risk of road traffic crashes, this study introduced risky driving behavior and traffic operating status and other big data, and extracted sharp acceleration, deceleration, turns, merge into other lane, traffic volume, average speed, and congestion index as variables. And traffic accident identification models were constructed based on accident data. The traffic accident identification model was evaluated based on the logistic regression algorithm. On the one hand, the contribution of risky driving behavior in traffic accident identification was quantified; on the other hand, the trend of traffic accident occurrence probability before and after the accident was analyzed. The results show that the sensitivity and AUC values of the traffic accident identification model considering both traffic operation state and risky driving behavior are increased by 5.00% and 0.03, respectively. The false alarm rate and missing report rate are decreased by 1.78% and 5.00%, respectively, which shows better fitting effect of the model. In addition, before and after the occurrence of traffic accidents, the risk probability of traffic crashes shows an obvious trend of rise, which is the key period of traffic accident prevention and control. The measures should be taken in corresponding sections in time to curb the rising trend of traffic crash risk and avoid the occurrence of traffic crashes. This study can provide intuitive basis for traffic accident prevention, and active prevention and control. Table and Figures | Reference | Related Articles | Metrics | Comments（0）

Select

Axial Compression Behavior of High-Strength Circular Steel Tube Confined High-Strength Steel Reinforced Concrete Short Columns KANG Lan, CHEN Zonglin, LIN Yiwei Journal of South China University of Technology(Natural Science Edition)    2022, 50 (7): 1-12.   DOI: 10.12141/j.issn.1000-565X.210463 Abstract （4153）   HTML （20）    PDF（pc） （2161KB）（112）       Knowledge map    Save This paper carried out axial compression tests on nine circular steel tube confined steel reinforced (CSTCSR) concrete short columns, one circular steel tube confined (CSTC) concrete short column and one circular steel tube column filled with steel-reinforced concrete. The main purpose of this research is to study the influences of yield strength of steel tube and shape steel, diameter-thickness ratio of steel tube, concrete strength, inner surface treatment method of high-strength steel tube and restraint mode on the failure mode, strain response, axial compression bearing capacity and ductility of CSTCSR concrete short column. The results show that the failure mode of high-strength circular steel tube confined high-strength steel reinforced (HCSTCHSR) concrete short column is overall shear failure, and there is no obvious local buckling on the surface of high-strength steel tube, and the development of oblique cracks of concrete is effectively restricted by the embedded high-strength shape steel. For CSTCSR concrete short column, the axial compression bearing capacity ratio and ductility coefficient of this column with high-strength steel tube and high-strength shape steel are increased from 1.37 to 1.49 and from 2.22 to 3.25, respectively, compared with those of CSTCSR concrete short column using ordinary-strength steel tube and ordinary-strength shape steel. It is concluded that the HCSTCHSR concrete short column has more excellent axial bearing capacity and ductility, and high-strength steel using in such column can be fully utilized. On the basis of “technical standard for steel tube confined concrete structures” (JGJ/T471—2019) and the results of parametric analysis in this study, a modified formula for axial bearing capacity of HCSTCHSR concrete short column was proposed by using the existing confined concrete’s constitutive model to provide scientific basis and data support for practical engineering application. Table and Figures | Reference | Related Articles | Metrics | Comments（0）

Select

Design of an Innovative Eddy Current Replaceable Coupling Beam and Its Numerical Analysis GONG Nan, LI Peizhen, HE Xuming Journal of South China University of Technology(Natural Science Edition)    2022, 50 (7): 25-34.   DOI: 10.12141/j.issn.1000-565X.210665 Abstract （3522）   HTML （9）    PDF（pc） （2320KB）（193）       Knowledge map    Save This paper carried out a detailed study on the damping characteristics of the eddy current coupling beam damper, which can start to dissipate energy under small deformation of the replaceable coupling beam. Based on the analysis of magnetic circuit theory, the study proposed the optimal arrangement of permanent magnet pole in eddy current damper. In other words, the permanent magnet poles parallel to the direction of conductor motion were arranged alternately, and the permanent magnet poles perpendicular to the direction of conductor motion were arranged in the same direction. In view of this, two kinds of eddy current dampers were designed, one of which is the plate eddy current damper with the conductor plate moving straight in the magnetic field and the other is the rotary eddy current damper with the gear-rack mechanism to amplify the rotation speed of the conductor plate in the magnetic field. Two kinds of eddy current dampers were used in the replaceable coupling beam, and the finite element simulation of the new eddy current coupling beam damper installed on the replaceable coupling beam was carried out, which revealed the nonlinear mechanical behavior of eddy current damping. It shows that the damping coefficient and stiffness coefficient are strongly related to the frequency. The higher the loading frequency, the lower the energy consumption efficiency and the higher the dynamic stiffness of the structure. So, the eddy current damper is more suitable for low frequency working conditions, and at this time, the damping coefficient of the eddy current damper is large, the energy consumption efficiency is high, and the stiffness coefficient is small, which basically does not change the natural vibration characteristics of the structure. Therefore, it is of great value in real-world application. Table and Figures | Reference | Related Articles | Metrics | Comments（0）

Select

Research on Roadside Environment Safety Based on Driver’s Attention Distribution Model WU Yanxia, ZHOU Tong, HUANG Shuai, et al Journal of South China University of Technology(Natural Science Edition)    2022, 50 (9): 49-57.   DOI: 10.12141/j.issn.1000-565X.20210802 Abstract （3283）   HTML （7）    PDF（pc） （11743KB）（145）       Knowledge map    Save In order to explore the influence of the highway roadside environment on the driver’s attention，optimize the roadside landscape and traffic sign settings，and improve the driving safety，this study determined the roadside environment risk level interval by quantitatively analyzing the driver’s attention distribution. Four types of typical scene were selected according to the actual state of the roadside environment， and field driving tests were carried out to collect data such as driver’s eye movement parameters and heart rate. The driver’s eye movement on the corresponding road section was analyzed，including fixation behavior，scan behavior and blinking behavior. On this basis，the obvious eye movement index that characterizes the driver’s attention was determined as the fixation parameter. According to the actual scene corresponding to the driver’s fixation point and the position of the point in the visual field， the driver's visual field area was divided into three parts， namely the roadside area （S area），the way area （W area） and the car area （C area）. By constructing the driver’s attention distribution model，the roadside environment was quantified and expressed by the ratio of attention region. The relationship model between the ratio of attention region and the driver’s heart rate growth rate was established，and the risk level interval was divided. The results show that the complexity of the roadside environment has a significant impact on the driver’s eye movement behavior. Compared with scan and blink，the driver’s fixation parameter can significantly characterize the driver’s attention state. When driving on the road，the driver’s fixation point moves in the view field，the ratio of attention region is the ratio of cumulative fixation time in the S area and the total fixation time of the S area and the W area. Its safety interval is ［9.93%，62.10%］，the risk interval is ［6.44%，9.93%） and （62.10%，76.93%］，and the danger interval is ［0，6.44%） and （76.93%，100%］. This research can provide a reference for the safety evaluation and decision-making of roadside landscape improvement. Table and Figures | Reference | Related Articles | Metrics | Comments（0）

Select

Trajectory Friction Compensation Algorithm for Robots Based on Velocity Control YE Bosheng, LI Siao, TAN Shuai, et al Journal of South China University of Technology(Natural Science Edition)    2024, 52 (4): 51-58.   DOI: 10.12141/j.issn.1000-565X.230368 Abstract （3260）   HTML （4）    PDF（pc） （2377KB）（51）       Knowledge map    Save Currently, robots are extensively utilized in industrial manufacturing. However, due to the influence of joint friction and other factors in the robot system, the robot trajectory tracking accuracy is difficult to meet the requirements of high-precision production. In this study, a friction compensation control algorithm in speed mode was proposed to mitigate the impact of non-linear friction factors in the mechanical structure and unmodelled disturbances on the robot’s operational stability and machining precision. The optimal excitation trajectory was designed by a combination of Fourier series and fifth-order polynomial. Dynamic parameters were then pre-identified by the least squares method and iteratively optimized through the Levenberg-Marquardt method to establish a more precise robot dynamic model. Subsequently, the Lyapunov method was adopted to design the trajectory tracking control algorithm, and the joint angles collected in the steepest discrete tracking differentiator were fed into the control algorithm to calculate the real-time compensation. The compensation value was then applied in the robot, which effectively achieving friction compensation. The proposed algorithm was validated by employing a six-degree-of-freedom serial robot as an experimental subject. The results demonstrate that the trajectory tracking error is reduced by approximately 35%, as comparing with that under the non-compensation conditions, which confirms the efficacy of the algorithm in the realm of robot friction compensation. Table and Figures | Reference | Related Articles | Metrics | Comments（0）

Select

Braking Collision Avoidance System for Vehicles Driving on Superhighway Based on Co-simulation HE Yongming, FENG Jia, QUAN Cong, et al Journal of South China University of Technology(Natural Science Edition)    2022, 50 (10): 19-28.   DOI: 10.12141/j.issn.1000-565X.210703 Abstract （3241）   HTML （6）    PDF（pc） （2693KB）（67）       Knowledge map    Save To solve the safety problem of collision between a high-speed intelligent vehicle and a low-speed vehicle on the superhighway, the vehicle braking collision avoidance system was studied by using the method of co-simulation. CarSim was used to establish the vehicle dynamics model and set the front vehicle parameters, road parameters and sensor parameters, and the control model based on vehicle distance and speed was established in MATLAB/Simulink. The signal connection is established through the input and output parameter interface module of CarSim software. When the speed of the front car is 100, 120 and 140 km/h, and the speed of an intelligent vehicle is 140, 160 and 180 km/h, respectively, the control model sends the braking deceleration signal to the intelligent vehicle through the real-time distance and speed collected by the sensor and establishes the emergency braking collision avoidance strategy for the vehicle on the superhighway. The results show that when the road adhesion coefficient is 0.60 and the car is braked on the flat straight section of the superhighway, the optimal wheel cylinder pressure is 7 MPa, and at this time, the braking distance of the car is 170.3 m at a speed of 160 km/h. The front car travels at a speed of 100, 120 and 140 km/h, and the smart car brakes at 140, 160 and 180 km/h, respectively, to the same speed as the car in front, requiring relative distances of 10.8, 10.7 and 10.5 m, respectively. When the road adhesion coefficient is 0.60, the vehicle speed of the front vehicle is 100, 120 and 140 km/h, respectively. When the initial cylinder pressure is 1 MPa and the intelligent vehicle braking decelerates to the same speed as the front vehicle, the distance between the front suspension of an intelligent vehicle and the rear suspension of the vehicle in front is 3.1, 3.5, and 3.8 m, respectively. When the initial cylinder pressure is 3 MPa and the intelligent vehicle braking decelerates to the same speed as the front vehicle, the distance between the front suspension of an intelligent vehicle and the rear suspension of the vehicle in front is 7.0, 7.3, and 7.7 m, respectively. Through the CarSim/Simulink co-simulation platform of vehicle emergency braking and collision avoidance control, the validity and accuracy of superhighway braking and collision avoidance model are verified, which can improve the safety of superhighway driving. Table and Figures | Reference | Related Articles | Metrics | Comments（0）

Select

Robot Collision Detection Based on External Torque Observer ZHANG Tie, CHEN Yijie, ZOU Yanbiao Journal of South China University of Technology(Natural Science Edition)    2024, 52 (3): 84-92.   DOI: 10.12141/j.issn.1000-565X.230086 Abstract （3192）   HTML （2）    PDF（pc） （2873KB）（40）       Knowledge map    Save Collision detection technology can reduce the probability of equipment damage and personal injury and plays an important role in modern human-robot collaborative production. To realize the collision detection without external torque sensor, it is necessary to accurately estimate the external torque of industrial robots. However, the accuracy of external torque estimation can be affected by parameters identification error of dynamic model and measurement error of motor current. To solve these problems, this paper designed a disturbance Kalman filter external torque observer based on the disturbance principle. The observer takes the equivalent external torque of external collision as the disturbance term, defines the joint disturbance model, and introduces the generalized momentum of the robot to construct the state-space equation. Considering the parameters identification error of the dynamic model and the measurement error of the motor current, it carried out an iterative estimation based on Kalman filter algorithm to obtain the optimal external torque. In order to improve the sensitivity of collision detection, a time-varying symmetric threshold function which varies with joint velocity was proposed for collision detection. The proposed method can adjust the threshold according to the change of joint velocity to adapt to the observed values of external torques at different working speeds. Experimental results show that compared with the generalized momentum observer, the accuracy of external torque estimation of the proposed observer is improved by 52.03%. In order to verify the effectiveness of the proposed method, this paper used a 6-DOF series joint industrial robot to conduct collision detection experiments. The experimental results show that compared with the static threshold, the time-varying threshold method reduces the detection delay by 58.06%, which can improve the sensitivity of collision detection and is more conducive to the safe operation and collision protection of industrial robots. Table and Figures | Reference | Related Articles | Metrics | Comments（0）

Select

Travel Carbon Emission Prediction Model Based on Resident Attribute Data SU Yuejiang, WEN Huiying, YUAN Minxian, WU Dexin, ZHOU Lulu, QI Weiwei Journal of South China University of Technology(Natural Science Edition)    2024, 52 (8): 23-33.   DOI: 10.12141/j.issn.1000-565X.230355 Abstract （3138）   HTML （7）    PDF（pc） （5415KB）（216）       Knowledge map    Save It is an important basis for precise formulation of transportation emission reduction measures to accurately analyze the importance of factors influencing residents’ travel mode and the sensitivity of carbon emissions. According to the comprehensive analysis of the influencing factors such as family attributes, personal attributes, travel attributes and environmental attributes of the residents’ travel survey, the prediction model of residents’ travel mode was constructed based on LightGBM (Light Gradient Boosting Machine) and verified. Combined with the travel activity level, the carbon emission coefficient of various energy types, the standard coal coefficient and other parameters, the travel carbon emission prediction model based on the resident attribute data was constructed. Finally, taking Guangzhou as an example, the carbon emission intensity and total amount of residents’ travel mode were predicted, and importance of factors influencing travel mode and sensitivity was analyzed.The results indicate that the carbon emission prediction model constructed based on the attribute data of residents can more accurately predict the carbon emission of various modes of travel, better analyze the importance and sensitivity of the influencing factors of carbon emission, and comprehensively reveal the relationship between travel behavior, travel mode and travel carbon emission. Among them, the distance between the start and the end and the nearest bus station or the distance from the nearest subway station, the cost of self-driving and travel distance are important factors affecting the choice of residents’ travel mode. The competitiveness of subway travel increases significantly with the decrease of distance when the distance between the starting and the end point and the nearest subway station drops by 55%. In the area with high density of bus stops, the distance between the start and the nearest bus station is not sensitive to residents’ travel mode choice. It is the turning point of the residents’ travel mode and carbon emission when the carbon emission cost increases by 400%. After passing the turning point, the car travel mode is difficult to transfer. The carbon emissions fell the fastest, with a maximum reduction of 90.4% when the reduction in travel distance was within 90%. Reference | Related Articles | Metrics | Comments（0）

Select

Lightweight Object Detection Combined with Multi-Scale Dilated-Convolution and Multi-Scale Deconvolution YI Qingming, LÜ Renyi, SHI Min, et al Journal of South China University of Technology(Natural Science Edition)    2022, 50 (12): 41-48.   DOI: 10.12141/j.issn.1000-565X.220095 Abstract （3082）   HTML （10）    PDF（pc） （2447KB）（210）       Knowledge map    Save Due to the tough issues of slow detection and heavy parameters, the deep neural networks are inapplicable to be deployed on mobile application scenarios which are computing-resource-constrained but demand high speed calculation. To improve the inference speed for object detection and achieve a better tradeoff between detection accuracy and inference speed, this paper proposed a lightweight object detection network named MDDNet which combined multi-scale dilated-convolution and multi-scale deconvolution. Firstly, a lightweight detection backbone network was designed based on an efficient single-stage strategy, and the depthwise separable convolution was introduced to reduce the parameter amount of the baseline and further speed up the feature extraction. Secondly, two feature extension branches based on multi-scale dilated convolution were added to the backbone network, which were respectively connected to the ends of the final and the penultimate residual layers of the basic network. The features of the two branches were fused in the prediction layer to augment the texture features of the shallow feature maps. Thirdly, the multi-scale deconvolution module was further introduced and connected to the deep feature network layers to increase the size of the feature map, and then the shallow feature maps of the previous layer with different scales were fused so as to enrich the feature semantic information and the detailed information, improving the detection accuracy. Finally, the parameters of the prior bounding box were optimized in the prediction layer based on the K-means clustering method, so that the prior bounding box could better match the ground truth of the object, achieving higher object recognition accuracy. The experimental results show that the MDDNet produces about 7.21×106 parameters. The average accuracy is 58.7% and 76.0% in KITTI and Pascal VOC datasets, respectively, while the corresponding inference speed respectively reaches 55 f/s and 52 f/s in the above two datasets. Therefore, MDDNet achieves a decent tradeoff among the parameter amount, detection speed, and detection accuracy, and it can be applied to real-time object detection on mobile terminals. Table and Figures | Reference | Related Articles | Metrics | Comments（0）

Select

Resilience Evolution of Multi-mode Transportation Network in Urban Agglomeration Based on Risk Diffusion MA Shuhong, YANG Lei, CHEN Xifang Journal of South China University of Technology(Natural Science Edition)    2023, 51 (6): 42-51.   DOI: 10.12141/j.issn.1000-565X.220486 Abstract （3019）   HTML （4）    PDF（pc） （4048KB）（106）       Knowledge map    Save It can help improve the risk resilience of urban agglomerations to carry out in-depth analysis of the dynamic resilience evolution characteristics of multi-modal transport networks in urban agglomerations during the risk diffusion phase. Based on complex network theory and its extension theory, this study constructed a multi-modal and multi-level transportation network model for urban agglomerations based on road, railway and air networks. It matched node degree and node betweenness with accessibility, and analyzed their static topological characteristics under risk diffusion. Based on the theory of cascading failure dynamics, it considered the initial risk level, risk warning threshold and risk resilience of nodes at different stages, and constructed a risk dynamic diffusion model based on the nodes and connected edge measures affecting risk diffusion. Considering the characteristics of structural and functional changes under risk shocks, the paper constructed a network structural and functional resilience measurement model, and the resilience performance of multi-modal transport networks was represented by their coupling values. Using the multi-modal transport network of the Guan-Zhong Plain urban agglomeration as the research object, the Python Networkx and Matlab network analysis tools were used to simulate and analyze the dynamic resilience evolution of the multi-modal transport network of the urban agglomeration during the risk diffusion phase with respect to different risk diffusion methods, network reliability, redundancy, robustness and node risk handling capacity coefficients. The results show that the model results are consistent with reality. The network resilience can be effectively enhanced by improving network reliability, redundancy, robustness and node risk handling. Compared to the risk diffusion approach based on the node measure, the risk diffusion approach based on the connected edge measure has greater impact on resilience performance, suggesting that the distribution of route hierarchy levels has a greater impact on the resilience performance of multimodal transport networks in urban agglomerations compared to the number of routes. The overall resilience of a multi-modal, multi-level transport network performs better than that of a single transport network. Table and Figures | Reference | Related Articles | Metrics | Comments（0）

Select

Fundamental Diagram Model of Mixed Traffic Flow of Connected Vehicles Considering Time Delay LUO Ruifa, HAO Huijun, XU Taorang, et al Journal of South China University of Technology(Natural Science Edition)    2023, 51 (1): 106-113.   DOI: 10.12141/j.issn.1000-565X.210702 Abstract （3013）   HTML （8）    PDF（pc） （2861KB）（100）       Knowledge map    Save Due to the continuous development of connected and automated vehicles, there will be a mixed traffic flow in which intelligent networked vehicles and manual driving vehicles coexist in the future. Therefore, the study of mixed traffic on the road can effectively solve traffic congestion and other problems, so it has certain practical significance. In order to explore the relationship between the flow, density and speed of this type of mixed traffic flow, this paper established a fundamental diagram model of the mixed traffic flow in an autonomous driving environment based on the comprehensive consideration of the degradation of intelligent networked vehicles and the delay between vehicles. First, it determined the types of vehicles in the traffic flow and the proportion of different types of vehicle, and considered the vehicle functional degradation when connected intelligent vehicles follow artificial vehicles. Then, the delay time of each type of three vehicles was determined and the following model of each vehicle was improved. On this basis, considering both the vehicle delay and the vehicle function degradation, the fundamental diagram model of traffic flow balance was derived, and the sensitivity analysis of free flow speed parameter in the model was carried out. The research result shows that connected and automated vehicles have a positive impact on the maximum flow and optimal density of mixed traffic flows, while vehicle delays have a negative impact; the free flow speed has a positive impact on the maximum flow and a negative impact on the optimal density of mixed traffic flow. The SUMO simulation results show that the simulated flow-density distribution points in different scenarios conform to the theoretical curve, which verifies the accuracy of the theoretical model in the paper. Table and Figures | Reference | Related Articles | Metrics | Comments（0）

Select

Metro Transfer Passenger Flow Prediction Based on STL-GRU ZHAO Jiandong, ZHU Dan, LIU Jiaxin Journal of South China University of Technology(Natural Science Edition)    2022, 50 (5): 22-31.   DOI: 10.12141/j.issn.1000-565X.210559 Abstract （2996）      PDF（pc） （1580KB）（437）       Knowledge map    Save A metro transfer passenger flow prediction model was proposed based on the seasonal decomposition of time series by loess（STL）and Gated Recurrent Unit（GRU），in order to enrich the research on metro internal transfer passenger flow prediction and to better formulate the metro operation plan.The prediction process was divided into three stages by the model.In the first stage，the raw automatic fare collection（AFC）data are preprocessed，where the travel path of passengers is identified using the graph-based depth-first search algorithm and the transfer passenger flow time series are constructed.In the second stage，the transfer passenger flow time series are decomposed into the trend component，seasonal component and remainder component by the STL；while the outliers of remainder component are eliminated and filled using the 3σ principle.In the third stage，the GRU model is built and the related training and prediction are processed through the deep learning library Keras.The model performance was validated with the passenger flow data of Xizhimen Station of Beijing metro.The result shows that，compared to the following 3 models which are long short-term memory neural network（LSTM），GRU and STL-LSTM model，the STL-GRU prediction model can improve the prediction accuracy of transfer passenger flow on weekdays（excluding Friday），Friday and weekends，and the mean absolute percentage errors of the prediction results can be reduced by at least 2.3%，1.36%，and 6.42%，respectively. Related Articles | Metrics | Comments（0）

Select

BiLSTM-BiDAF Named Entity Recognition Based on Machine Reading Comprehension WANG Jie, XIA Xiaoming Journal of South China University of Technology(Natural Science Edition)    2022, 50 (12): 80-88.   DOI: 10.12141/j.issn.1000-565X.220013 Abstract （2901）   HTML （11）    PDF（pc） （1576KB）（98）       Knowledge map    Save Named entity recognition is a fundamental task of natural language processing (NLP) and plays an important role in many downstream NLP tasks, including information extraction and machine translation, etc. The existing named entity recognition methods are usually based on sequence labeling and extract entities within a sentence independently. These methods ignore the semantic information between sentences. Named entity recognition methods based on machine reading comprehension encode important prior information about the entity class. It is easier to distinguish similar classification labels, which reduces the difficulty of model learning, but it still only models at the sentence level, ignoring the semantic information between sentences, which is easy to cause the problem of inconsistent entity labeling in different sentences. To this end, this paper extended the sentence-level named entity recognition to the text-level named entity recognition, and then proposed a BiLSTM-BiDAF named entity recognition model based on machine reading comprehension. First, to utilize the context information within the whole text, NEZHA pre-training language model was used to obtain information of the full text and local features were further captured through BiLSTM, so as to strengthen the model’s ability to capture locally dependent information. Then, a bidirectional attention flow was introduce to learn the semantic association between the text and entity category. Finally, to predict the position of entities in the text, a boundary detector based on the gating mechanism was design to strengthen the correlation of the entity boundary. At the same time, an answer count detector was establish to identify the unanswerable questions. Experimental results on the CCKS2020 Chinese electronic medical records dataset and CMeEE dataset show that our model can effectively identify document-level and sentence-level named entities, and F1 can reach 84.76% and 57.35%, respectively. Table and Figures | Reference | Related Articles | Metrics | Comments（0）

Select

Connection Scheme Optimization of Last Trains of Urban Mass Transit Network Based on Considering the Transfer Passengers ZHENG Yajing, LI Yaohui, JIN Wenzhou Journal of South China University of Technology(Natural Science Edition)    2022, 50 (5): 32-39.   DOI: 10.12141/j.issn.1000-565X.210567 Abstract （2840）      PDF（pc） （749KB）（67）       Knowledge map    Save Focusing on the transfer connection problem of last trains of urban mass transit network，this paper analyzed the complexity of the connection relationship of last trains of urban mass transit network and established an optimization model for the connection scheme of last trains of urban mass transit aiming at maximizing the number of passengers.The essence of the model is to solve the maximum directed acyclic subgraph of the weighted directed graph.And then，an appropriate coding method was designed，the ox like method was used for cross operation，and a genetic algorithm suitable for the optimization model of last trains connection scheme of urban mass transit was proposed.Finally，the proposed genetic algorithm was verified with an example.The result shows that the algorithm can quickly obtain a more optimized last trains connection scheme，which is easy to be realized by computer.This method can be used as an auxiliary means for the preparation of last trains schedule，and provides a certain decision-making basis for the preparation of last trains schedule of each line in the urban mass transit network. Related Articles | Metrics | Comments（0）

Select

Analysis on Propagation of Spatio-Temporal Dynamic Effects Towards Freeway Traffic Crash YANG Yang, HU Yanran, YUAN Zhenzhou, et al Journal of South China University of Technology(Natural Science Edition)    2023, 51 (1): 123-133.   DOI: 10.12141/j.issn.1000-565X.220353 Abstract （2772）   HTML （14）    PDF（pc） （2051KB）（201）       Knowledge map    Save This paper aims to explore and quantify the spatio-temporal propagation mechanism of freeway traffic crashes and address the defect of traditional traffic wave theory considering single dimension in such problems. Firstly, the speed changing before and after the traffic crash was analyzed based on the dynamic traffic flow data collected by microwave radar detectors. Furthermore, the speed changing rate was introduced as a measure of the impact of a crash, the bilinear interpolation method was adopted to construct the speed changing rate curve, and the Savitzky-Golay filtering fitting method was applied to fit the outer contour of the spatio-temporal region affected by the crash. Finally, the indicators of spatio-temporal impacts of traffic crash were figured out and analyzed under the condition that the speed changing rate threshold is 20%, 30% and 40%, respectively. The indicators include the beginning time of the crash effect, the end time of the crash effect, the duration of the crash effect, the closest distance of the crash effect, the furthest distance of the crash effect, the spatial range of the crash effect, the propagation speed of the crash effect and the dissipation speed of crash effect. The results indicate that the smaller the threshold of speed changing rate is, the earlier the crash influence starts, the later the crash influence ends, the longer the duration and the longer the distance of the crash influence has; the higher the speed changing rate threshold is, the later the crash influence starts, the earlier the crash influence ends, the shorter the duration and the shorter the distance of the crash influence have; additionally, under various speed changing rate threshold, the changing trend of propagation speed generated by the traffic crash varies with time. The approach adopted in this paper has strong operability and high identification of results, which can provide theoretical support for real-time traffic control and guidance after freeway crashes. Table and Figures | Reference | Related Articles | Metrics | Comments（0）