摘要
室内环境下,针对惯性定位中传感器通过周期采样控制器采样频率高,造成采样信号的冗余,控制器节点对这些采样信号进行不必要计算,导致计算资源浪费的问题,现提出视觉惯性融合定位下的事件触发机制,构造事件触发控制器。首先建立室内四旋翼的动力学模型,在ROS中模拟视觉惯性融合定位算法,通过实验平台验证。其次以滚转角系统为例,用反步法设计控制器,并用Lyapunov稳定性理论证明系统稳定性。实验结果显示:在室内特征不明显情况下,飞行器在实际飞行过程中有较好的准确性,且3s内周期采样控制器采样3000次,所设计的事件触发仅需触发150次就可得到相同的跟踪效果,即此设计的事件触发控制器减少了飞行器内部的计算量,极大的节约了通讯资源。
In indoor environment,the sampling frequency of sensors passing through the periodic sampling control-ler in inertial positioning is high,which leads to the redundancy of sampling signals,and the unnecessary calculation of these sampling signals by the controller nodes leads to the waste of computing resources.In this paper,an event triggering mechanism based on visual inertial fusion positioning is proposed,and an event triggering controller is con-structed.Firstly,the dynamic model of indoor quadrotor was established,and the visual inertial fusion positioning algo-rithm was simulated in ROS,which was verified by the experimental platform.Secondly,taking the roll angle system as an example,the controller was designed by backstepping method,and the stability of the system was proved by Lya-punov stability theory.The experimental results show that the aircraft has good accuracy in the actual flight process when the indoor characteristics are not obvious,and the periodic sampling controller samples 3,000 times in 3s.The designed event trigger only needs to trigger 150 times to get the same tracking effect,that is,the designed event trigger controller reduces the internal calculation of the aircraft and greatly saves the communication resources.
作者
邓志良
袁玉雯
DENG Zhi-liang;YUAN Yu-wen(School of Automation,Nanjing University of Information Science&Technology,Nanjing Jiangsu 210044,China)
出处
《计算机仿真》
北大核心
2023年第11期22-26,81,共6页
Computer Simulation
基金
国家自然科学基金(51875293)
电子信息控制重点实验室开放基金项目(6142105190308)。
关键词
姿态跟踪控制
视觉惯性融合
事件触发机制
事件触发控制
Attitude tracking control
Visual inertial fusion
Event triggering mechanism
Event triggered control
