To overcome the shortcomings of the traditional passive ranging technology based on image, such as poor ranging accuracy, low reliability and complex system, a new visual passive ranging method based on re-entrant coa...To overcome the shortcomings of the traditional passive ranging technology based on image, such as poor ranging accuracy, low reliability and complex system, a new visual passive ranging method based on re-entrant coaxial optical path is presented. The target image is obtained using double cameras with coaxial optical path. Since there is imaging optical path difference between the cameras, the images are different. In comparison of the image differences, the target range could be reversed. The principle of the ranging method and the ranging model are described. The relationship among parameters in the ranging process is analyzed quantitatively. Meanwhile,the system composition and technical realization scheme are also presented. Also, the principle of the method is verified by the equivalent experiment. The experimental results show that the design scheme is correct and feasible with good robustness. Generally, the ranging error is less than 10% with good convergence. The optical path is designed in a re-entrant mode to reduce the volume and weight of the system. Through the coaxial design,the visual passive range of the targets with any posture can be obtained in real time. The system can be widely used in electro-optical countermeasure and concealed photoelectric detection.展开更多
The vehicles and pedestrians ranging is one of the basic functions of advanced driving assistance system.However,most of the ranging systems can only work on workstations with high computing power.To solve this proble...The vehicles and pedestrians ranging is one of the basic functions of advanced driving assistance system.However,most of the ranging systems can only work on workstations with high computing power.To solve this problem,a lightweight algorithm is proposed to be packaged into Android application package,and be installed in Android smartphones for vehicles and pedestrians ranging.The proposed ranging system is based on the images obtained by smartphone’s monocular camera.To achieve real-time ranging,an 8-bit integer(int8)quantization algorithm is proposed to accelerate the inference of convolutional neural networks.To increase the detection precision,a zoom-in algorithm is further proposed to detect small targets in the distance.After having detected the 2D bounding boxes of vehicles and pedestrians,a pinhole ranging method is applied to estimate the distance.In order to verify the proposed algorithm,the mean average precision(mAP)and the frame per second(FPS)are first tested by using COCO dataset on Huawei P40Pro,then,the ranging precision on the real road.The experimental results show that this algorithm can successfully perform real-time ranging(15 FPS)with high precision(34.8 mAP)onto the tested smartphones.Finally,a possible mobile application based on the ranging algorithm,i.e.,distance keeping warning,is also provided.展开更多
Exit choice is one of the most important pedestrian behaviors during evacuation.Distance to the exit is a generally recognized factor influencing expected moving time to the exit.Visual range determines how much infor...Exit choice is one of the most important pedestrian behaviors during evacuation.Distance to the exit is a generally recognized factor influencing expected moving time to the exit.Visual range determines how much information a pedestrian can perceive,thus the number of pedestrians within the visual field can be used to estimate waiting time at the exit.Besides,the choice firmness that reflects the degree to which a pedestrian would persist in his/her previous choice of exit is proposed.By integrating game theory into a cellular automata simulation framework,the pedestrian exit choice mechanism is investigated and explicitly modeled in this paper.A systematic analysis of the key factors influencing pedestrian evacuation is conducted,including visual radius and choice firmness of a pedestrian,initial crowd distribution of the room,exit layout as well as exit width.It is found that low choice firmness level can lead to unnatural pedestrian behavior such as wandering,which is adverse to evacuation.The longer the pedestrian's visual radius,the earlier the pedestrian can determine his/her final selection of the exit.Compared with the scenario where the pedestrians are randomly distributed,pedestrians clustered together in a corner of the room lead to high crowd density and imbalanced use of exits.Furthermore,the exit layout and exit width also have a certain influence on pedestrian evacuation process.The results of this paper may be of benefit to the formulation of behavioral rules in other pedestrian simulation models.展开更多
To develop the dynamic monitoring algorithm of visual safety distance in highway, by using the highway video traffic monitoring system, the research platform of four kinds of terrain environment in plateau, mountainou...To develop the dynamic monitoring algorithm of visual safety distance in highway, by using the highway video traffic monitoring system, the research platform of four kinds of terrain environment in plateau, mountainous area, plain and coastal area is established. Results show that through the contrast between the sample data and visibility train of thought, based on the theory of mathematical morphology, expressway visibility dynamic monitoring image information system can be established. Based on the theory of the measurement of the basic formula of visibility, the dynamic model of the optimization is established, set up 200 meters distance visual observation target system, research visual range detection algorithm process.展开更多
Based on the hourly observation data of heavy rainfall at Changbei Airport from 2011 to 2018,the main characteristics of heavy rainfall at the airport were analyzed. The results show that the heavy rainfall at Changbe...Based on the hourly observation data of heavy rainfall at Changbei Airport from 2011 to 2018,the main characteristics of heavy rainfall at the airport were analyzed. The results show that the heavy rainfall at Changbei Airport often occurred from April to August,most in June,mostly from the evening to midnight. Visibility and runway visual range( RVR) were often affected by the heavy rainfall,but the proportion of weather processes below the take-off and landing standards was relatively small. The main influencing factors of heavy rainfall were low vortex shear,front,high trough and subtropical high edge. The changing characteristics of meteorological elements in the high-centroid rainfall processes were more obvious,and they had greater impact on flight.展开更多
Modern beyond visual range air game is characterized by high dynamic complexity and multi-factor interweaving,making it challenging for traditional decision-making methods.This paper constructs an autonomous decision-...Modern beyond visual range air game is characterized by high dynamic complexity and multi-factor interweaving,making it challenging for traditional decision-making methods.This paper constructs an autonomous decision-making framework for a single-plane beyond visual range air game that encompasses the entire game process.We design tactical rules with both offensive and defensive capabilities and formulate an autonomous air game decision-making method.This method is primarily driven by rules and is complemented by an intelligent maneuver decision-making approach based on the Double Deep Q-Network(DDQN).Simulation results demonstrate that our designed method effectively enhances the maneuvering decision-making ability of aircraft and has promising application prospects.展开更多
基金Supported by the National Basic Research Program of China under Grant No 2014CB340102
文摘To overcome the shortcomings of the traditional passive ranging technology based on image, such as poor ranging accuracy, low reliability and complex system, a new visual passive ranging method based on re-entrant coaxial optical path is presented. The target image is obtained using double cameras with coaxial optical path. Since there is imaging optical path difference between the cameras, the images are different. In comparison of the image differences, the target range could be reversed. The principle of the ranging method and the ranging model are described. The relationship among parameters in the ranging process is analyzed quantitatively. Meanwhile,the system composition and technical realization scheme are also presented. Also, the principle of the method is verified by the equivalent experiment. The experimental results show that the design scheme is correct and feasible with good robustness. Generally, the ranging error is less than 10% with good convergence. The optical path is designed in a re-entrant mode to reduce the volume and weight of the system. Through the coaxial design,the visual passive range of the targets with any posture can be obtained in real time. The system can be widely used in electro-optical countermeasure and concealed photoelectric detection.
基金the RoboCar Project for Internationalization of RTI Projects within the Frame of the Austrian Research Promotion Agency(No.861000)。
文摘The vehicles and pedestrians ranging is one of the basic functions of advanced driving assistance system.However,most of the ranging systems can only work on workstations with high computing power.To solve this problem,a lightweight algorithm is proposed to be packaged into Android application package,and be installed in Android smartphones for vehicles and pedestrians ranging.The proposed ranging system is based on the images obtained by smartphone’s monocular camera.To achieve real-time ranging,an 8-bit integer(int8)quantization algorithm is proposed to accelerate the inference of convolutional neural networks.To increase the detection precision,a zoom-in algorithm is further proposed to detect small targets in the distance.After having detected the 2D bounding boxes of vehicles and pedestrians,a pinhole ranging method is applied to estimate the distance.In order to verify the proposed algorithm,the mean average precision(mAP)and the frame per second(FPS)are first tested by using COCO dataset on Huawei P40Pro,then,the ranging precision on the real road.The experimental results show that this algorithm can successfully perform real-time ranging(15 FPS)with high precision(34.8 mAP)onto the tested smartphones.Finally,a possible mobile application based on the ranging algorithm,i.e.,distance keeping warning,is also provided.
基金the National Natural Science Foundation of China(Grant No.71904116)the Fund from the Shanghai Municipal Commission of Science and Technology,China(Grant Nos.19DZ1209600 and 18DZ1201500)。
文摘Exit choice is one of the most important pedestrian behaviors during evacuation.Distance to the exit is a generally recognized factor influencing expected moving time to the exit.Visual range determines how much information a pedestrian can perceive,thus the number of pedestrians within the visual field can be used to estimate waiting time at the exit.Besides,the choice firmness that reflects the degree to which a pedestrian would persist in his/her previous choice of exit is proposed.By integrating game theory into a cellular automata simulation framework,the pedestrian exit choice mechanism is investigated and explicitly modeled in this paper.A systematic analysis of the key factors influencing pedestrian evacuation is conducted,including visual radius and choice firmness of a pedestrian,initial crowd distribution of the room,exit layout as well as exit width.It is found that low choice firmness level can lead to unnatural pedestrian behavior such as wandering,which is adverse to evacuation.The longer the pedestrian's visual radius,the earlier the pedestrian can determine his/her final selection of the exit.Compared with the scenario where the pedestrians are randomly distributed,pedestrians clustered together in a corner of the room lead to high crowd density and imbalanced use of exits.Furthermore,the exit layout and exit width also have a certain influence on pedestrian evacuation process.The results of this paper may be of benefit to the formulation of behavioral rules in other pedestrian simulation models.
文摘To develop the dynamic monitoring algorithm of visual safety distance in highway, by using the highway video traffic monitoring system, the research platform of four kinds of terrain environment in plateau, mountainous area, plain and coastal area is established. Results show that through the contrast between the sample data and visibility train of thought, based on the theory of mathematical morphology, expressway visibility dynamic monitoring image information system can be established. Based on the theory of the measurement of the basic formula of visibility, the dynamic model of the optimization is established, set up 200 meters distance visual observation target system, research visual range detection algorithm process.
基金Supported by Meteorological Science and Technology Project of Jiangxi Province(17302005151714)。
文摘Based on the hourly observation data of heavy rainfall at Changbei Airport from 2011 to 2018,the main characteristics of heavy rainfall at the airport were analyzed. The results show that the heavy rainfall at Changbei Airport often occurred from April to August,most in June,mostly from the evening to midnight. Visibility and runway visual range( RVR) were often affected by the heavy rainfall,but the proportion of weather processes below the take-off and landing standards was relatively small. The main influencing factors of heavy rainfall were low vortex shear,front,high trough and subtropical high edge. The changing characteristics of meteorological elements in the high-centroid rainfall processes were more obvious,and they had greater impact on flight.
文摘Modern beyond visual range air game is characterized by high dynamic complexity and multi-factor interweaving,making it challenging for traditional decision-making methods.This paper constructs an autonomous decision-making framework for a single-plane beyond visual range air game that encompasses the entire game process.We design tactical rules with both offensive and defensive capabilities and formulate an autonomous air game decision-making method.This method is primarily driven by rules and is complemented by an intelligent maneuver decision-making approach based on the Double Deep Q-Network(DDQN).Simulation results demonstrate that our designed method effectively enhances the maneuvering decision-making ability of aircraft and has promising application prospects.
