基于虚拟图像的乡村公园互动景观装置系统的设计研究  

Design and Research of Interactive Landscape Installation System for Rural Parks Based on Virtual Images

作　　者：张晨[1] ZHANG Chen(Xi’an International University,Xi’an 710077,China)

机构地区：[1]西安外事学院,西安710077

出　　处：《自动化与仪器仪表》2024年第5期106-109,共4页Automation & Instrumentation

摘　　要：为了构建一个基于虚拟图像的观光车驾驶模拟系统,此次研究基于三维虚拟图像,设计一个虚拟图像标定样板对双目视角和物体图文视认进行标定。基于观光车动力学系统,采用快速扩展随机树算法对行驶路径进行规划。结果表明,标定的数值与实际测量的数值误差分别为3.57%、2.70%、2.68%、4.42%、1.30%、2.60%和1.30%,误差都在5%以内,满足了标定的允许误差范围。经过标定样板标定后的虚拟场景和真实场景结果接近。改进后的快速扩展随机树算法经过300次迭代后趋于稳定,规划路线耗时6.8 s,平均路线长度为105 m。改进后的RRT算法在经过220次迭代后趋于收敛,规划路线总耗时4.0 s,平均路线长度为132 m。以上结果证明了标定样板标定虚拟图像以及改进快速扩展随机树算法对路线规划的有效性。In order to construct a simulation system for sightseeing vehicle driving based on virtual images,this study designs a virtual image calibration template based on 3D virtual images to calibrate binocular perspectives and object image recognition.Based on the dynamic system of sightseeing vehicles,a fast expanding random tree algorithm is used to plan the travel path.The results show that the errors between the calibrated values and the actual measured values are 3.57%,2.70%,2.68%,4.42%,1.30%,2.60%,and 1.30%,respectively,all within 5%,meeting the allowable error range for calibration.The results of the virtual scene calibrated by the calibration template are close to those of the real scene.The improved fast expanding random tree algorithm stabilized after 300 iterations,took 6.8 seconds to plan the route,and the average route length was 105 meters.The improved RRT algorithm tends to converge after 220 iterations,with a total planning time of 4.0 seconds and an average route length of 132 meters.The above results demonstrate the effectiveness of calibrating virtual images with calibration templates and improving the fast expanding random tree algorithm for route planning.

关 键 词：虚拟图像 公园 互动 观光 快速扩展随机树 

分 类 号：TU856[建筑科学] TP273[自动化与计算机技术—检测技术与自动化装置]