杠杆(统计)
控制理论(社会学)
计算机科学
扰动(地质)
避障
控制器(灌溉)
空气动力学
模型预测控制
可靠性(半导体)
控制工程
控制(管理)
工程类
机器人
人工智能
航空航天工程
移动机器人
古生物学
功率(物理)
物理
量子力学
农学
生物
作者
Dadong Fan,Qianyuan Liu,Chenlong Zhao,Kexin Guo,Ziqi Yang,Xiang Yu,Lei Guo
出处
期刊:IEEE robotics and automation letters
日期:2024-05-15
卷期号:9 (7): 6328-6335
标识
DOI:10.1109/lra.2024.3401127
摘要
Safe and autonomous flight of quadrotors in enclosed environments still remains formidable challenge due to the aerodynamic proximity effect and restricted free space. This letter develops an integrated planning and control architecture for disturbance-aware and safety control of quadrotors. By explicitly considering ground effect in both the controller and obstacle avoidance constraints, the proposed method can fully leverage the benefits brought by aerodynamic disturbance. Specifically, a Ground Effect Observer (GEO) is designed to estimate repulsive force based on the ground effect model. Subsequently, building upon the augmented quadrotor model, a Nonlinear Model Predictive Controller (NMPC) is developed to enhance tracking performance while conserving energy during flight. Moreover, aided by the Exponential Control Barrier Functions (ECBFs), the distance-based safety constraints can be mapped onto the constraints of control inputs. This facilitates the consideration of disturbance on safety constraints, allowing quadrotors to avoid obstacles promptly. Experimental evaluation employs two types of tunnels to assess the applicability of the proposed scheme. Comparative results show that the proposed scheme outperforms several popular methods in safety and reliability.
科研通智能强力驱动
Strongly Powered by AbleSci AI