Joint Optimization of Transmit Power and Trajectory for UAV-Enabled Data Collection With Dynamic Constraints

• Published in: IEEE transactions on communications Vol. 73; no. 9; pp. 8080 - 8091
• Main Authors: Zhang, Hongyun, Li, Bin, Rong, Yue, Zeng, Yong, Zhang, Rui
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.09.2025; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Autonomous aerial vehicles; Constraints; control parametrization; Control theory; Data collection; Differential equations; Electronic mail; Internet of Things; Mathematical models; Nonlinear dynamical systems; Optimization; Parameterization; Penalty function; Performance degradation; power allocation; State space models; State-space methods; Trajectory; trajectory optimization; UAV-enabled data collection; Unmanned aerial vehicles; Vehicle dynamics
• ISSN: 0090-6778; 1558-0857
• DOI: 10.1109/TCOMM.2025.3543221

The unmanned aerial vehicle (UAV)-enabled data collection system with a rotary-wing UAV and multiple ground nodes (GNs) is investigated in this paper. The average transmission data rate is maximized through the coordinated optimization of the GNs' transmit power and the UAV's trajectory. In particular, the UAV dynamic constraints and physical constraints are imposed. The UAV dynamics, which are governed by a group of differential equations, are usually ignored in existing works. As a consequence, the planned trajectory cannot be fully tracked by the controller in real world applications, which could lead to severe performance degradation. Thus, a control-based method is devised to address this issue. Specifically, by adopting the state-space model from control theory, the data collection problem is established as a dynamic optimization problem subject to state constraints, in which both of the decision variables and constraints are infinite-dimensional in nature. The key idea of the solution method is to convert the infinite-dimensional dynamic program into a finite-dimensional static nonlinear problem. This is achieved by deriving the required gradients of the dynamic optimization problem based on the control parametrization scheme and an exact penalty function method. The effectiveness and superiority of the proposed design are validated via numerical experiments.