The numerical simulations and experimental results of an ultra-broadband acoustic ventilation barrier composed of periodic unit cells are reported in this paper.Based on multiple mechanisms,including sound absorption ...The numerical simulations and experimental results of an ultra-broadband acoustic ventilation barrier composed of periodic unit cells are reported in this paper.Based on multiple mechanisms,including sound absorption by eigenmodes of the unit cell and sound reflection by a plate structure on upper surface of the unit cell,a single-layer ventilation barrier with broadband sound reduction is designed,and its working bandwidth can reach about 1560 Hz.The experimental results accord well with the simulation results.Furthermore,two types of three-layer ventilation barriers are designed and demonstrated by using the unit cells with different values of a(the length of the hollow square region)and w(the width of the channel between the adjacent cavities),and the bandwidths of both ventilation barriers can increase to 3160 Hz and 3230 Hz,respectively.The designed barrier structures have the advantages of ultra-broadband sound reduction and ventilation,which paves the way to designing high-performance ventilation barriers for the applications in environmental protection and architectural acoustics.展开更多
目的激光雷达实时定位与建图(simultaneous localization and mapping,SLAM)是智能机器人领域的重要组成部分,通过对周边环境的3维建模,可以实现无人驾驶车辆的自主定位和精准导航。针对目前单个车辆激光雷达建图周期长、算力需求大的现...目的激光雷达实时定位与建图(simultaneous localization and mapping,SLAM)是智能机器人领域的重要组成部分,通过对周边环境的3维建模,可以实现无人驾驶车辆的自主定位和精准导航。针对目前单个车辆激光雷达建图周期长、算力需求大的现状,提出了基于边缘计算的多车协同建图方法,能够有效地负载均衡,在保证单个车辆精准定位的同时,增加多个车辆之间的地图重用性。方法构建基于阈值的卸载函数,论证边缘计算下的多车卸载决策属于势博弈问题,设计实现基于边缘计算的势博弈卸载算法,在模型具有纳什均衡的基础上实现任务调度,引入α-Nash最佳响应动态加速算法收敛,并采用由粗到细的点云匹配方法提高地图匹配性能,实现车辆的精准定位。最后,基于地图的相对可信度,高效地合并基站覆盖范围内的多个车辆的建图数据。结果实验表明,基于博弈论的调度方法在保证定位可靠性的前提下,能够有效地实现多车协同SLAM,且多车协同的定位与建图结果与使用载波相位差分技术(real-time kinematic,RTK)的高精度差分全球定位系统(differential global positioning system,DGPS)结果足够接近,相比于单车建图而言,横向定位和纵向定位的平均精度分别提高了6.0倍和3.9倍。结论本文方法解决了基于边缘计算的多车协同激光雷达SLAM问题,借助边缘服务器的计算资源,无人驾驶车辆可以有效地减少本地资源需求和定位延迟。该方法通过各个车辆之间的资源博弈,最终实现纳什均衡。实现基于边缘计算的激光雷达定位服务,且高效地完成多车之间的地图合并,仿真和真实环境中的实验表明了方法的有效性。展开更多
基金Project supported by the National Natural Science Foundation of China (Grant Nos.12174159,12274183,and 51976079)the National Key Research and Development Program of China (Grant No.2020YFC1512403)the Research Project of State Key Laboratory of Mechanical System and Vibration (Grant No.MSV202201)。
文摘The numerical simulations and experimental results of an ultra-broadband acoustic ventilation barrier composed of periodic unit cells are reported in this paper.Based on multiple mechanisms,including sound absorption by eigenmodes of the unit cell and sound reflection by a plate structure on upper surface of the unit cell,a single-layer ventilation barrier with broadband sound reduction is designed,and its working bandwidth can reach about 1560 Hz.The experimental results accord well with the simulation results.Furthermore,two types of three-layer ventilation barriers are designed and demonstrated by using the unit cells with different values of a(the length of the hollow square region)and w(the width of the channel between the adjacent cavities),and the bandwidths of both ventilation barriers can increase to 3160 Hz and 3230 Hz,respectively.The designed barrier structures have the advantages of ultra-broadband sound reduction and ventilation,which paves the way to designing high-performance ventilation barriers for the applications in environmental protection and architectural acoustics.
文摘目的激光雷达实时定位与建图(simultaneous localization and mapping,SLAM)是智能机器人领域的重要组成部分,通过对周边环境的3维建模,可以实现无人驾驶车辆的自主定位和精准导航。针对目前单个车辆激光雷达建图周期长、算力需求大的现状,提出了基于边缘计算的多车协同建图方法,能够有效地负载均衡,在保证单个车辆精准定位的同时,增加多个车辆之间的地图重用性。方法构建基于阈值的卸载函数,论证边缘计算下的多车卸载决策属于势博弈问题,设计实现基于边缘计算的势博弈卸载算法,在模型具有纳什均衡的基础上实现任务调度,引入α-Nash最佳响应动态加速算法收敛,并采用由粗到细的点云匹配方法提高地图匹配性能,实现车辆的精准定位。最后,基于地图的相对可信度,高效地合并基站覆盖范围内的多个车辆的建图数据。结果实验表明,基于博弈论的调度方法在保证定位可靠性的前提下,能够有效地实现多车协同SLAM,且多车协同的定位与建图结果与使用载波相位差分技术(real-time kinematic,RTK)的高精度差分全球定位系统(differential global positioning system,DGPS)结果足够接近,相比于单车建图而言,横向定位和纵向定位的平均精度分别提高了6.0倍和3.9倍。结论本文方法解决了基于边缘计算的多车协同激光雷达SLAM问题,借助边缘服务器的计算资源,无人驾驶车辆可以有效地减少本地资源需求和定位延迟。该方法通过各个车辆之间的资源博弈,最终实现纳什均衡。实现基于边缘计算的激光雷达定位服务,且高效地完成多车之间的地图合并,仿真和真实环境中的实验表明了方法的有效性。