Aim To find an effective method to remove the registration error in multi-sensor systems. Methods A Kalman filtering technique was proposed to estimate and remove sensor bias and sensor fare tilt errors in multisenso...Aim To find an effective method to remove the registration error in multi-sensor systems. Methods A Kalman filtering technique was proposed to estimate and remove sensor bias and sensor fare tilt errors in multisensor systems with a moving platform. Results Simulation results are presented to demonstrate the performance of the approach. Conclusion The Kalman filter algorithm am detect registration errors and use this information to converge tracks from independent sensors. This is particularly important if the data from the sensors are to fused.展开更多
Focusing on the vibration of the roadbed and ground induced by high-speed train load, a three dimensional finite element model which includes the roadbed and horizontal layered site is established to study how the sit...Focusing on the vibration of the roadbed and ground induced by high-speed train load, a three dimensional finite element model which includes the roadbed and horizontal layered site is established to study how the site conditions, the load moving speed and the depth of the soil element influence the soil element stress response. Based on a track-subsoil analytical model in which the rail is simulated as an Euler-Bernoulli beam resting on Winkler foundation in the vertical plane, the reaction force between the sleeper and roadbed excited by a single axle is presented, and then that is exerted on relevant elements to simulate the moving load. The dynamic response in the roadbed and subsoil excited by a single axle moving load is computed based on the parallel computing platform of the ABAQUS finite element software, and the stress time-history, stress path and curves of the principal stress axes rotation of the soil element under the track are presented. The results show that： the soil element stress path is an apple-shaped curve in the horizontal shear stress τd versus the stress difference （σsh - σch ）/2 coordinate system; the principal stress axes rotate 180° for the soil element under the load moving line during the load running, and the stress state changes from the pure shear to triaxial shear and then back to the pure shear again. The element dynamic stress increases as the moving load speed increases, which increases sharply when the load speed approaches the Rayleigh wave velocity of soil layer; the site conditions and the soil element depth affect the soil element stress path significantly.展开更多
In order to solve the problem of small object size and low detection accuracy under the unmanned aerial vehicle(UAV)platform,the object detection algorithm based on deep aggregation network and high-resolution fusion ...In order to solve the problem of small object size and low detection accuracy under the unmanned aerial vehicle(UAV)platform,the object detection algorithm based on deep aggregation network and high-resolution fusion module is studied.Furthermore,a joint network of object detection and feature extraction is studied to construct a real-time multi-object tracking algorithm.For the problem of object association failure caused by UAV movement,image registration is applied to multi-object tracking and a camera motion discrimination model is proposed to improve the speed of the multi-object tracking algorithm.The simulation results show that the algorithm proposed in this study can improve the accuracy of multi-object tracking under the UAV platform,and effectively solve the problem of association failure caused by UAV movement.展开更多
文摘Aim To find an effective method to remove the registration error in multi-sensor systems. Methods A Kalman filtering technique was proposed to estimate and remove sensor bias and sensor fare tilt errors in multisensor systems with a moving platform. Results Simulation results are presented to demonstrate the performance of the approach. Conclusion The Kalman filter algorithm am detect registration errors and use this information to converge tracks from independent sensors. This is particularly important if the data from the sensors are to fused.
文摘Focusing on the vibration of the roadbed and ground induced by high-speed train load, a three dimensional finite element model which includes the roadbed and horizontal layered site is established to study how the site conditions, the load moving speed and the depth of the soil element influence the soil element stress response. Based on a track-subsoil analytical model in which the rail is simulated as an Euler-Bernoulli beam resting on Winkler foundation in the vertical plane, the reaction force between the sleeper and roadbed excited by a single axle is presented, and then that is exerted on relevant elements to simulate the moving load. The dynamic response in the roadbed and subsoil excited by a single axle moving load is computed based on the parallel computing platform of the ABAQUS finite element software, and the stress time-history, stress path and curves of the principal stress axes rotation of the soil element under the track are presented. The results show that： the soil element stress path is an apple-shaped curve in the horizontal shear stress τd versus the stress difference （σsh - σch ）/2 coordinate system; the principal stress axes rotate 180° for the soil element under the load moving line during the load running, and the stress state changes from the pure shear to triaxial shear and then back to the pure shear again. The element dynamic stress increases as the moving load speed increases, which increases sharply when the load speed approaches the Rayleigh wave velocity of soil layer; the site conditions and the soil element depth affect the soil element stress path significantly.
基金the National Natural Science Foundation of China (No.61627810)the National Science and Technology Major Program of China (No.2018YFB1305003)the National Defense Science and Technology Outstanding Youth Science Foundation (No.2017-JCJQ-ZQ-031)。
文摘In order to solve the problem of small object size and low detection accuracy under the unmanned aerial vehicle(UAV)platform,the object detection algorithm based on deep aggregation network and high-resolution fusion module is studied.Furthermore,a joint network of object detection and feature extraction is studied to construct a real-time multi-object tracking algorithm.For the problem of object association failure caused by UAV movement,image registration is applied to multi-object tracking and a camera motion discrimination model is proposed to improve the speed of the multi-object tracking algorithm.The simulation results show that the algorithm proposed in this study can improve the accuracy of multi-object tracking under the UAV platform,and effectively solve the problem of association failure caused by UAV movement.
