Criteria

Text:
Sector:
Display:

Results

Viewing 241 to 270 of 85301
2017-10-08
Technical Paper
2017-01-2196
Giuseppe Cicalese, Fabio Berni, Stefano Fontanesi, Alessandro D'Adamo, Enrico Andreoli
Abstract High power-density Diesel engines are characterized by remarkable thermo-mechanical loads. Therefore, compared to spark ignition engines, designers are forced to increase component strength in order to avoid failures. 3D-CFD simulations represent a powerful tool for the evaluation of the engine thermal field and may be used by designers, along with FE analyses, to ensure thermo-mechanical reliability. The present work aims at providing an integrated in-cylinder/CHT methodology for the estimation of a Diesel engine thermal field. On one hand, in-cylinder simulations are fundamental to evaluate not only the integral amount of heat transfer to the combustion chamber walls, but also its point-wise distribution. To this specific aim, an improved heat transfer model based on a modified thermal wall function is adopted to estimate correctly wall heat fluxes due to combustion.
2017-10-08
Technical Paper
2017-01-2197
Vignesh Pandian Muthuramalingam, Anders Karlsson
Abstract Owing to increased interest in blended fuels for automotive applications, a great deal of understanding is sought for the behavior of multicomponent fuel sprays. This sets a new requirement on spray model since the volatility of the fuel components in a blend can vary substantially. It calls for careful solution to implement the differential evaporation process concerning thermodynamic equilibrium while maintaining a robust solution. This work presents the Volvo Stochastic Blob and Bubble (VSB2) spray model for multicomponent fuels. A direct numerical method is used to calculate the evaporation of multicomponent fuel droplets. The multicomponent fuel model is implemented into OpenFoam CFD code and the case simulated is a constant volume combustion vessel. The CFD code is used to calculate liquid penetration length for surrogate diesel (n-dodecane)-gasoline (iso-octane) blend and the result is compared with experimental data.
2017-10-08
Technical Paper
2017-01-2198
Zhihong Li, Guoxiu Li, Lan Wang, Hongmeng Li, Jie Wang, Haizhou Guo, Shuangyi He
Abstract The electromagnetic valve driving mechanism is the significant equipment, which plays a vital role in the unit pump injection system; therefore, the performance of the electromagnetic valve directly influences the function of the control system. Based on the operation conditions of the unit pump injection system, a steady electromagnetic valve model was modified to study the influence factors of electromagnetic force and the best combination to get the maximum electromagnetic force. The validation model was verified by experiment. The effects of some crucial parameters upon the electromagnetic force were investigated in the present paper, (including working airspace, magnetic pole’s cross-sectional area, coil position, coil turn, the armature thickness). The results show that the electromagnetic force of the solenoid valve enhanced with the increase of driving current and with the decrease of working airspace.
2017-10-08
Technical Paper
2017-01-2199
Maria Cristina Cameretti, Vincenzo De Bellis, Luca Romagnuolo, Agostino Iorio, Luigi Maresca
Abstract In recent years, engine manufacturers have been continuously involved in the research of proper technical solutions to meet more and more stringent CO2 emission targets, defined by international regulations. Many strategies have been already developed, or are currently under study, to attain the above objective. A tendency is however emerging towards more innovative combustion concepts, able to efficiently burn lean or highly diluted mixtures. To this aim, the enhancement of turbulence intensity inside the combustion chamber has a significant importance, contributing to improve the burning rate, to increase the thermal efficiency, and to reduce the cyclic variability. It is well-known that turbulence production is mainly achieved during the intake stroke. Moreover, it is strictly affected by the intake port geometry and orientation.
2017-10-08
Technical Paper
2017-01-2200
Peter Priesching, Mijo Tvrdojevic, Ferry Tap, Casper Meijer
Abstract Turbulent combustion modeling in a RANS or LES context imposes the challenge of closing the chemical reaction rate on the sub-grid level. Such turbulent models have as their two main ingredients sources from chemical reactions and turbulence-chemistry interaction. The various combustion models then differ mainly by how the chemistry is calculated (level of detail, canonical flame model) and on the other hand how turbulence is assumed to affect the reaction rate on the sub-grid level (TCI - turbulence-chemistry interaction). In this work, an advanced combustion model based on tabulated chemistry is applied for 3D CFD (computational fluid dynamics) modeling of Diesel engine cases. The combustion model is based on the FGM (Flamelet Generated Manifold) chemistry reduction technique. The underlying chemistry tabulation process uses auto-ignition trajectories of homogeneous fuel/air mixtures, which are computed with detailed chemical reaction mechanisms.
2017-10-08
Technical Paper
2017-01-2201
Zhongye Cao, Tianyou Wang, Kai Sun, Lei Cui, Yong Gui
Abstract For uniflow scavenged two-stroke marine diesel engines, the main function of scavenging process is to replace the burned gas with fresh charge. The end state of scavenging process is integral to the subsequent compression and combustion, thereby affecting the engine’s fuel economy, power output and emissions. In this paper, a complete working cycle of a large marine diesel engine was simulated by using the 3D-CFD software CONVERGE. The model was validated by mesh sensitivity test and experiment data. Based on this calibrated model, the influences of swirl ratio and exhaust valve closing (EVC) timing on the scavenging process were investigated. The parameters evaluating the performance of scavenging process were introduced. The results show that, by adjusting the swirl orientation angle(SOA) from SOA=10° to SOA=30°, different swirl ratios are generated and have obvious differences in flow characteristics and scavenging performance.
2017-10-08
Technical Paper
2017-01-2202
Shiyou Yang
Abstract This work presents an application of two sub-models relative to chemical-kinetics-based turbulent pre-mixed combustion modeling approach on the simulation of burn rate and emissions of spark ignition engines. In present paper, the justification of turbulent pre-mixed combustion modeling directly based on chemical kinetics plus a turbulence model is given briefly. Two sub-models relative to this kind of pre-mixed combustion modeling approach are described generally, including a practical PRF (primary reference fuel) chemical kinetic mechanism which can correctly capture the laminar flame speed under a wide range of Ford SI (spark ignition) engines/operating conditions, and an advanced spark plug ignition model which has been developed by Ford recently.
2017-10-08
Technical Paper
2017-01-2205
Velmurugan M A, MahendraMohan Rajagopal
Abstract Agricultural tractors are often subjected to various applications like front end loading work, cultivation work, where frequent forward and reverse gears are needed. Most of Indian agricultural tractors are equipped with mechanical transmission system which demands repeated clutching and de-clutching operation for such applications resulting in increased operator fatigue and lesser productivity. Also need of electronics in Indian agricultural industry for better farm mechanization is growing high. This research work depicts development of electronic bi-directional shifting (power shuttle) control design and calibration for farm vehicle fitted with wet clutch transmission. This research also reduces operator fatigue via frequent directional shift through electronic transmission. The control system is designed without any electronic interfacing with engine and also provides clutch-less gear shifting and auto-launch which offers ease to drive even for novice driver.
2017-10-08
Technical Paper
2017-01-2204
Hoon Lee, Kwangwoo Jeong, Sanghoon Yoo, Byungho Lee, Sejun Kim
Abstract Hyundai-Kia Motor Company recently developed a multi-way, electrical coolant valve for engine thermal management module (TMM). The main purposes of the TMM are to boost fuel economy by accelerating engine warm-up and also to enhance engine thermal efficiency by actively controlling the operating temperature. In addition to those, the system can improve vehicle heating and cooling performance as well. The electrical coolant valve is a key component in the TMM as it modulates the amount of coolant flow to individual components in cooling system such as engine oil heat exchanger, heater core, and radiator. The coolant flow modulation is done by controlling the electric valve’s position with using an electric motor attached to the valve. The objective of the valve control is to manage coolant temperature at a desired level that varies depending on vehicle’s operating condition. This paper discusses the control algorithm developed for controlling electrical coolant valve.
2017-10-08
Technical Paper
2017-01-2208
Tao Liu, Ziwang Lu, Guangyu Tian
Abstract To further explore the potential of fuel economy for hybrid electric vehicle (HEV), a methodology of demand power optimization is proposed. The fuel consumption depends not only on the EMS, but also on the way to operate vehicle. A control strategy to adjust driver’s demand before power splitting is necessary. To get accurate and reliable control strategy, two aspects are the most important. First, a rigorous and organized modeling approach is a base to describe complicated powertrain system of HEV. The energetic macroscopic representation (EMR) is a graphical synthetic description of electromechanical conversion system based on energy flow. A powertrain architecture of HEV is described explicitly via the EMR. Second, the effectiveness of EMS and the reasonability of driving operations are vital.
2017-10-08
Technical Paper
2017-01-2207
Nobunori Okui
Abstract Next-generation vehicles which include the Electric Vehicles, the Hybrid Electric Vehicles and the Plug-in Hybrid Electric Vehicles are researched and expected to reduce carbon dioxide (CO2) emission in the future. In order to reduce the emissions of the heavy-duty diesel plug-in hybrid electric vehicles (PHEV), it is necessary to provide the high exhaust-gas temperature and to keep the exhaust-gas aftertreatment system effective. The engine starting condition of the PHEV is cold, and the engine start and stop is repeated. And, the engine load of the PHEV is assisted by the electric motor. Therefore, the exhaust-gas aftertreatment system of the PHEV is not able to get the enough high exhaust-gas temperature. And, the warm-up of the exhaust-gas aftertreatment system for the PHEV is spent the long time. So, it is worried about a bad effect on the emission characteristics of the PHEV.
2017-10-08
Journal Article
2017-01-2434
Srinivasan Paulraj, Saravanan Muthiah
Traditionally driveline ratios are selected based on trial and error method of proto vehicle testing. This consumes lot of time and increases overall vehicle development effort. Over last few decades, simulation-based design approach has been extensively used to alleviate this problem. This paper describes torque converter and final drive ratio (FDR) selection at concept phase for new Automatic Transmission (AT) vehicle development. Most of the critical data required for simulating vehicle performance and fuel economy (FE) targets were not available (e.g. shift map, clutch slip map, pedal map, dynamic torque, coast down, etc.) at an initial stage of the project. Hence, the risk for assuming right inputs and properly selecting FDR/Torque converter was particularly high. Therefore, a validated AVL Cruise simulation model based on an existing AT vehicle was used as a base for new AT vehicle development to mitigate the risk due to non-availability of inputs.
2017-10-08
Technical Paper
2017-01-2369
Prakash Arunachalam, Martin Tuner, Per Tunestal, Marcus Thern
Abstract Humid air motor (HAM) is an engine operated with humidified inlet charge. System simulations study on HAM showed the waste heat recovery potential over a conventional system. An HAM setup was constructed, to comprehend the potential benefits in real-time, the HAM setup was built around a 13-litre six cylinder Volvo diesel engine. The HAM engine process is explained in detail in this paper. Emission analysis is also performed for all three modes of operation. The experiments were carried out at part load operating point of the engine to understand the effects of humidified charge on combustion, efficiency, and emissions. Experiments were conducted without EGR, with EGR, and with humidified inlet charge. These three modes of operation provided the potential benefits of each system. Exhaust heat was used for partial humidification process. Results show that HAM operation, without compromising on efficiency, reduces NOx and soot significantly over the engine operated without EGR.
2017-10-08
Technical Paper
2017-01-2209
Christian Ibron, Mehdi Jangi, Tommaso Lucchini, Xue-Song Bai
Partially premixed combustion (PPC) can be applied to decrease emissions and increase fuel efficiency in direct injection, compression ignition (DICI) combustion engines. PPC is strongly influenced by the mixing of fuel and oxidizer, which for a given fuel is controlled mainly by (a) the fuel injection, (b) the in-cylinder flow, and (c) the geometry and dynamics of the engine. As the injection timings can vary over a wide range in PPC combustion, detailed knowledge of the in-cylinder flow over the whole intake and compression strokes can improve our understanding of PPC combustion. In computational fluid dynamics (CFD) the in-cylinder flow is sometimes simplified and modeled as a solid-body rotation profile at some time prior to injection to produce a realistic flow field at the moment of injection. In real engines, the in-cylinder flow motion is governed by the intake manifold, the valve motion, and the engine geometry.
2017-10-08
Technical Paper
2017-01-2210
Masayoshi Matsuda, Takeshi Yokomori, Norimasa Iida
Abstract The thermal efficiency of a spark-ignition (SI) engine must be improved to reduce both environmental load and fuel consumption. Although lean SI engine operation can strongly improve thermal efficiency relative to that of stoichiometric SI operation, the cycle-to-cycle variation (CCV) of combustion increases with the air dilution level. Combustion CCV is caused by CCVs of many factors, such as EGR, spark energy, air-fuel ratio, and in-cylinder flow structure related to engine speed. This study focuses on flow structures, especially the influence of a tumble structure on flow fluctuation intensity near ignition timing. We measured the flow field at the vertical center cross section of an optically accessible high-tumble flow engine using time-resolved particle image velocimetry.
2017-10-08
Technical Paper
2017-01-2211
Mengqiu Jia, Zhen Lu, Tianyou Wang, Yufeng Li, Yanzhe Sun, Ming Wen, Zhizhao Che, Kai Sun
Abstract The intake process plays an important role in the operation of internal combustion engines. In the present study, a three-dimensional transient simulation of a four-valve diesel engine was performed using Large Eddy Simulation (LES) model based on software CONVERGE. The mean velocity components in three directions through the intake valve curtain, the flow separation around the intake valves, the influences of inlet jet on turbulence flow field and cycle-to-cycle variation were investigated in this work. The result shows that the mean velocity distributes non-uniformly near the valve curtain at high valve lifts. In contrast, the mean velocity distribution is uniform at low valve lifts. It is found that the flow separation occurs at valve stem, valve seat and valve sealing through the outlet of the helical port. In contrast, flow separation is only observed in the valve seat through the outlet of the tangential port.
2017-10-08
Technical Paper
2017-01-2219
Xihui Wang
Conventional heat transfer fluids used for cooling fluids in vehicle cooling water jacket have relatively poor heat transfer performance.One method for enhance heat transfer in cooling-jacket uses nanofluids. Nanofluids have heat transfer enhancement merits .In the present study, the numerical simulation on Fe3O4 nanofluid flow in cooling water jacket of Gasline direct injection engine wax performed using computational fluid dynamics ( CFD) software FLUENT. The heat transfer coefficient of nanofluids was calculated and verified by experiment. Fe3O4 nano-particles were used in mixture of water/ethylene glycol as a base fluid. The thermal performance of the nanofluid was studied, also the thermal performance of a cooling-jacket was studied with CFD software. The simulation was performed for different volumetric concentrations of(1%,2%,5%) nanofluids of different engine speeds.
2017-09-29
Technical Paper
2017-01-5011
Raviprakash Shankaranarayana
Process Parameters play a vital role in product quality of Injection Molded components. Variation in process parameters will lead to Injection Molded manufacturing defects like Sink Mark, Flow Mark, Silver Streak, Flash, Warping, Weld lines, Jetting, voids, Short Shot & Bubbles. The variation of process parameters can be optimized using Six Sigma approach which is a structured methodology which is Process focused & data driven approach. The purpose of this paper is to present through a case study how the concepts of Design of Experiments, which is a part of Six Sigma Methodology can be used for improving the Injection Molding Process at supplier end reducing defects & hence improving Quality. Here one of the products in Motorcycle Industry has been taken which has 100% Sink Mark defect & resulting in 100% rework.
2017-09-23
Journal Article
2017-01-1960
Xiaopeng Zong, Guoyan Xu, Guizhen Yu, Hongjie Su, Chaowei Hu
Abstract Obstacle avoidance is an important function in self-driving vehicle control. When the vehicle move from any arbitrary start positions to any target positions in environment, a proper path must avoid both static obstacles and moving obstacles of arbitrary shape. There are many possible scenarios, manually tackling all possible cases will likely yield a too simplistic policy. In this paper reinforcement learning is applied to the problem to form effective strategies. There are two major challenges that make self-driving vehicle different from other robotic tasks. Firstly, in order to control the vehicle precisely, the action space must be continuous which can’t be dealt with by traditional Q-learning. Secondly, self-driving vehicle must satisfy various constraints including vehicle dynamics constraints and traffic rules constraints. Three contributions are made in this paper.
2017-09-23
Journal Article
2017-01-1970
Guizhen Yu, Zhangyu Wang, Xinkai Wu, Yalong Ma, Yunpeng Wang
Abstract: In this paper, an efficient lane detection using deep feature extraction method is proposed to achieve real-time lane detection in diverse road environment. The method contains three main stages: 1) pre-processing, 2) deep lane feature extraction and 3) lane fitting. In pre-processing stage, the inverse perspective mapping (IPM) is used to obtain a bird's eye view of the road image, and then an edge image is generated using the canny operator. In deep lane feature extraction stage, an advanced lane extraction method is proposed. Firstly, line segment detector (LSD) is applied to achieve the fast line segment detection in the IPM image. After that, a proposed adaptive lane clustering algorithm is employed to gather the adjacent line segments generated by the LSD method. Finally, a proposed local gray value maximum cascaded spatial correlation filter (GMSF) algorithm is used to extract the target lane lines among the multiple lines.
2017-09-23
Technical Paper
2017-01-1951
Lingfei Wu, Hongshan Zha, Caijing Xiu, Qiaojun He
Abstract Local path planning for obstacle avoidance is one of the core topics of intelligent vehicle. A novel method based on dubins curve and tentacle algorithm is proposed in this article, with the consideration of obstacle avoidance and vehicle motion constraints. First, the preview distance of the vehicle is given according to the current speed, so that the preview point can be found with the information of global path. Then dubins curve is adopted to find a path with appropriate turning radius, between the current position and preview point, satisfying the constraints of current direction and target direction, considering handling and ride comfort of the vehicle. In order to avoid obstacle, tentacle algorithm is adopted. 20 tentacle points are given by moving the original preview point, and then 21 local paths can be given by using dubins curve. Cost function is used to find out the best option of the 21 paths.
2017-09-23
Technical Paper
2017-01-1954
Peng Hang, Xinbo Chen, Fengmei Luo
Abstract Path tracking is the rudimentary capability and primary task for autonomous ground vehicles (AGVs). In this paper, a novel four-wheel-independent-steering (4WIS) and four-wheel-independent-drive (4WID) electric vehicle (EV) is proposed which is equipped with steer-by-wire (SBW) system. For path-tracking controller design, the nonlinear vehicle model with 2 degrees of freedom (DOF) is built utilizing the nonlinear Dugoff tire model. The nonlinear dynamic model of SBW system is conducted as well considering the external disturbances. As to the path-tracking controller design, an integrated four-wheel steering (4WS) and direct yaw-moment control (DYC) system is designed based on the model predictive control (MPC) algorithm to track the target path described by desired yaw angle and lateral displacement. Then, the fast terminal sliding mode controller (FTSMC) is proposed for the SBW system to suppress disturbances.
2017-09-23
Technical Paper
2017-01-1955
Yandong Ruan, Hui Chen, Jiancong Li
Abstract An integrated automatic driving system consists of perception, planning and control. As one of the key components of an autonomous driving system, the longitudinal planning module guides the vehicle to accelerate or decelerate automatically on the roads. A complete longitudinal planning module is supposed to consider the flexibility to various scenarios and multi-objective optimization including safety, comfort and efficiency. However, most of the current longitudinal planning methods can not meet all the requirements above. In order to satisfy the demands mentioned above, a new Potential Field (PF) based longitudinal planning method is presented in this paper. Firstly, a PF model is constructed to depict the potential risk of surrounding traffic entities, including obstacles and roads. The shape of each potential field is closely related to the property of the corresponding traffic entity.
2017-09-23
Technical Paper
2017-01-1952
ChengJun Ma, Fang Li, Chenglin Liao, Lifang Wang
Abstract With the load of urban traffic system becomes more serious, the Automatic Parking System (APS) plays an important role in alleviating the burden of drivers and improving vehicle safety. The APS is consisted of environmental perception, path planning and path following. The path following controls the lateral movement of vehicle during the parking process, and requires the trajectory tracking error to be as small as possible. At present, some control algorithms are used including PID control, pure pursuit control, etc. However, these algorithms relying heavily on parameters and environment, have some problems such as slow response and low precision. To solve this problem, a path following control method based on Model Predictive Control (MPC) algorithm is proposed in this paper. Firstly, Kinematic vehicle model and path tracker based on MPC algorithm are built. Secondly, a test bench that composed of CANoe hardware in the loop (HIL) system and steering wheel system is built.
2017-09-23
Technical Paper
2017-01-1953
Manfei Bai, Lu Xiong, Zhiqiang Fu, Renxie Zhang
Abstract In this paper, a speed tracking controller is designed for the All-terrain vehicles. The method of feedforward with state variable feedback based on conditional integrators is adopted by the proposed control algorithm. The feedforward is designed considering the influence of the road slope on the longitudinal dynamics, which makes the All-terrain vehicles satisfy the acceleration demand of the upper controller when it tracks the desired speed on the road with slope varying greatly. The road slope is estimated based on a combined kinematic and dynamic model. This method solves the problem that road slope estimation requires an accurate vehicle dynamic model and are susceptible to acceleration sensor bias. Based on the vehicle dynamic model and the nonlinear tire model, the method of conditional integration is used in the state variable feedback, which considers the saturation constraint of the actuator with the intention of preventing the divergent integral operation.
2017-09-23
Technical Paper
2017-01-1963
Yuande Jiang, Weiwen Deng, Rui He, Shun Yang, Shanshan Wang, Ning Bian
Abstract Adaptive cruise control (ACC), as one of the advanced driver assistance systems (ADAS), has become increasingly popular in improving both driving safety and comfort. Since the objectives of ACC can be multi-dimensional, and often conflict with each other, it is a challenging task in its control design. The research presented in this paper takes ACC control design as a constrained optimization problem with multiple objectives. A hierarchical framework for ACC control is introduced, aimed to achieve optimal performance on driving safety and comfort, speed and/or distance tracking, and fuel economy whenever possible. Under the hierarchical framework, the operational mode is determined in the upper layer, in which a model predictive control (MPC) based spacing controller is employed to deal with the multiple control objectives. On the other hand, the lower layer is for actuator control, such as braking and driving control for vehicle longitudinal dynamics.
2017-09-23
Technical Paper
2017-01-1964
Xiangkun He, Xuewu Ji, Kaiming Yang, Yulong Liu, Jian WU, Yahui Liu
Abstract Highway traffic safety has been the most serious problem in current society, statistics show that about 70% to 90% of accidents are caused by driver operational errors. The autonomous emergency braking (AEB) is one of important vehicle intelligent safety technologies to avoid or mitigate collision. The AEB system applies the vehicle brakes when a collision is eminent in spite of any reaction by the driver. In some technologies, the system forewarns the driver with an acoustic signal when a collision is still avoidable, but subsequently applies the brakes automatically if the driver fails to respond. This paper presents the development and implementation of a rear-end collision avoidance system based on hierarchical control framework which consists of threat assessment layer, wheel slip ratio control layer and integrated-electro-hydraulic brake (IEHB) actuator control layer.
2017-09-23
Technical Paper
2017-01-1958
Dongfang Dang
Abstract With the increasing complexity, dynamicity and uncertainty of traffic, motion planning of automatic driving is getting more difficult and challenging. This paper focuses on the real-time motion planning problem of CAVs (connected and automated vehicles) in complex traffic scenarios. To effectively solve this problem, a general driving risk model is presented, which contains the following two essential parts: i) collision risk, i.e., the collision risk between the SV (subject vehicle) and other surrounding vehicles, pedestrians, buildings etc.; ii) non-collision risk, such as violation of traffic regulations, the deviation from the intention of driver, etc. To achieve the real time collision detection, the SV is approximated to a point and its shape is considered by extending the dimension of obstacles considering their relative position and velocity.
2017-09-23
Technical Paper
2017-01-1962
Hongluo Li, Yutao Luo
Abstract The trajectory planning and the accurate path tracking are the two key technologies to realize the intelligent driving. The research of the steering wheel angle plays an important role in the path tracking. The purpose of this study is to optimize the steering wheel angle input during the automated lane changing. A dynamic programming approach to trajectory planning is proposed in this study, which is expected to not only achieve a quick reaction to the changing driving environment, but also optimize the balance between vehicle performance and driving efficiency. First of all, the lane changing trajectory is planned based on the positive and negative trapezoidal lateral acceleration method. In addition, the multi-objective optimization function is built which includes such indexes: lateral acceleration, lateral acceleration rate, yaw rate, lane changing time and lane changing distance.
2017-09-23
Journal Article
2017-01-1969
Yuanxin Zhong, Sijia Wang, Shichao Xie, Zhong Cao, Kun Jiang, Diange Yang
Abstract Real-time reconstruction of 3D environment attributed with semantic information is significant for a variety of applications, such as obstacle detection, traffic scene comprehension and autonomous navigation. The current approaches to achieve it are mainly using stereo vision, Structure from Motion (SfM) or mobile LiDAR sensors. Each of these approaches has its own limitation, stereo vision has high computational cost, SfM needs accurate calibration between a sequences of images, and the onboard LiDAR sensor can only provide sparse points without color information. This paper describes a novel method for traffic scene semantic segmentation by combining sparse LiDAR point cloud (e.g. from Velodyne scans), with monocular color image. The key novelty of the method is the semantic coupling of stereoscopic point cloud with color lattice from camera image labelled through a Convolutional Neural Network (CNN).
Viewing 241 to 270 of 85301