• Journal of Information Processing Systems
• /
• v.17 no.2
• /
• pp.411-425
• /
• 2021

In this paper, a vehicle recognition algorithm based on deep convolutional neural network and compression dictionary is proposed. Firstly, the network structure of fine vehicle recognition based on convolutional neural network is introduced. Then, a vehicle recognition system based on multi-scale pyramid convolutional neural network is constructed. The contribution of different networks to the recognition results is adjusted by the adaptive fusion method that adjusts the network according to the recognition accuracy of a single network. The proportion of output in the network output of the entire multiscale network. Then, the compressed dictionary learning and the data dimension reduction are carried out using the effective block structure method combined with very sparse random projection matrix, which solves the computational complexity caused by high-dimensional features and shortens the dictionary learning time. Finally, the sparse representation classification method is used to realize vehicle type recognition. The experimental results show that the detection effect of the proposed algorithm is stable in sunny, cloudy and rainy weather, and it has strong adaptability to typical application scenarios such as occlusion and blurring, with an average recognition rate of more than 95%.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.33 no.4
• /
• pp.201-208
• /
• 2010

This research is about developing an efficient solution procedure for the vehicle routing problem under varying vehicle moving speeds for hour-based time interval. Different moving speeds for every hour is too difficult condition to solve for this type of combinatorial optimization problem. A methodology to divide the 12 hour based time interval offered by government into 5 different time intervals and then divide delivery area into 12 small divisions first and then re-organizing them into 5 groups. Then vehicle moving speeds are no longer varying in each of the 5 divisions. Therefore, a typical TSP solution procedure may be applied to find the shortest path for all 5 divisions and then connect the local shortest paths to form a delivery path for whole area. Developed solution procedures are explained in detail with 60 points example.

• Journal of Mechanical Science and Technology
• /
• v.15 no.12
• /
• pp.1639-1646
• /
• 2001

It is often necessary that the joints characteristics should be determined in the early stage of the vehicle body design. The researches on identification of joints in a vehicle body have been performed until the recent year. In this study, the joint characteristics of vehicle structure were expressed as the condensed matrix forms from the full joint stiffness matrix. The condensed joint stiffness matrix was applied to typical T-type and Edge-type joints, and the usefulness was confirmed. In addition, it was applied to the real center pillar model and the full vehicle body in order to validate the practical application.

• Advances in aircraft and spacecraft science
• /
• v.4 no.4
• /
• pp.437-448
• /
• 2017

The problem of flow through the vent is formulated as an unsteady, nonlinear, ordinary differential equation and solved using Runge-Kutta method to obtain pressure inside payload faring. An inverse problem for prediction of the discharge coefficient is presented employing measured internal pressure of the payload fairing during the ascent phase of a satellite launch vehicle. A controlled random search method is used to estimate the discharge coefficient from the measured transient pressure history during the ascent period of the launch vehicle. The algorithm predicts the discharge coefficient stepwise with function of Mach number. The estimated values of the discharge coefficients are in good agreement with differential pressure measured during the flight of typical satellite launch vehicle.

• Journal of Auto-vehicle Safety Association
• /
• v.6 no.2
• /
• pp.23-28
• /
• 2014

Commercial vehicle ESC assessment for curvature road was conducted. The previous study of ESC activation condition for losing controllability utilizing the test protocols of double lane change and sine with dwell method was conducted without considering the geometric complexity of roadway design. Since critical rollover accidents were frequently observed in the exit ramp zone, variety of curve, slope and bank have been added for analysis conditions. Detailed feature of the ramp including location, dimension and design characteristics have been analyzed from the typical trumpet type ramp design. Analyzing accident data from 2008, two specific ramps have been selected due to their complexity in design and severity in steering operation.

• Proceedings of the Korean Society for Agricultural Machinery Conference
• /
• 2000.11b
• /
• pp.325-332
• /
• 2000

The objective of this research work is to develop an autonomous levee-weeding robot. In this paper, pathway control system for the robot is developed and simulated. A prototype autonomous vehicle for levee weeding is also developed and used in the actual test. The results obtained in this research work is summarized as follows; 1) The simulated typical time history of lateral displacements and heading angle of the vehicle in straight run shows that the vehicle tendency is always to achieve the target path from any of its deviated position and heading angle. 2) The test run on an asphalt surface by the prototype crawler-type vehicle is in good agreement with the simulation results.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.6 no.4
• /
• pp.9-16
• /
• 1998

The joint characteristics are necessary to be determined in the early stage of the vehicle body design. Researches on identification of joints in a vehicle body have been performed until the recent year. In this study, the joint characteristics of vehicle struct- ure were expressed as condensed forms from the full joint stiffness and mass matrix. The condensed joint stiffness and mass matrix were applied to typical T-type and Edge-type joints, and the usefulness was confirmed. In addition, those were applied to center pillar and full vehicle body to validate the practical application.

• Proceedings of the Computational Structural Engineering Institute Conference
• /
• 2001.10a
• /
• pp.143-150
• /
• 2001

In this paper, the effects of vehicle loads on flat slab system are investigated based on the previous studies on beam-girder parking structural system. The influence surfaces of flat slab for typical design section are determined for the purpose of obtaining member forces under vehicle loads. In addition, the equivalent vehicle load factors for flat slab parking structures are suggested using neural network. It has been found that vehicle load effects of flat slab system are dominant for the center positive moment in both column and middle strip, as like beam-girder parking structural system.

• Journal of Power System Engineering
• /
• v.1 no.1
• /
• pp.106-123
• /
• 1997

In this study, vehicle directional stability is investigated for limit driving for crash avoidance maneuver using a full vehicle dynamic model. The model was analytically validated using typical step steering and lane change simulation. Limit driving condition for the vehicle model was quoted from research results of references. It was demonstrated that instable vehicle motion was caused by not only road conditions but also driving conditions. Also, the simulation showed that braking combined with steering caused very hazardous situation in crash avoidance maneuver. Finally, phase plane plot approach was used to evaluate the dynamic instability.

• Proceedings of the Computational Structural Engineering Institute Conference
• /
• 2002.04a
• /
• pp.233-240
• /
• 2002

In this paper, the effects of vehicle loads on flat slab system are investigated on the basis of the previous studies for beam-girder parking structural system. The influence surfaces of flat slab for typical design section are developed for the purpose of obtaining maximum member forces under vehicle loads. In addition, the equivalent vehicle load factors for flat slab parking structures are suggested using artificial neural network. The network responses are compared with the results by numerical analyses to verify the validation of Levenberg-Marquardt algorithm adopted as training method in this paper. Many parameter studies fur the flat slab structural system show dominant vehicle load effects at the center positive moments in both column and middle strips, like the beam-girder parking structural system.