Modeling and Control of Unmanned Aerial Vehicles – Current Status and Future Directions

Recent military and civil actions worldwide have highlighted the potential utility for Unmanned Aerial Vehicles (UAVs). Both fixed wing and rotary aircraft have contributed significantly to the success of several military and surveillance/rescue operations. Future combat operations will continue to place unmanned aircraft in challenging conditions such as the urban warfare environment. However, the poor reliability, reduced autonomy and operator workload requirements of current unmanned vehicles present a roadblock to their success. It is anticipated that future operations will require multiple UAVs performing in a cooperative mode, sharing resources and complementing other air or ground assets. Surveillance and reconnaissance tasks that rely on UAVs require sophisticated modeling, planning and control technologies. This paper reviews the current status of UAV technologies with emphasis on recent developments aimed at UAV improved autonomy and reliability and discusses future directions and technological challenges that must be addressed in the immediate future. We view the assembly of multiple and heterogeneous vehicles as a “system of systems” where individual UAVs are functioning as sensors or agents. Thus, networking, computing and communications issues must be considered as the UAVs are tasked to perform surveillance and reconnaissance missions in an urban environment. The same scenario arises in similar civil applications such as forest fire detection, rescue operations, pipeline monitoring, etc. A software (middleware) platform enables real time reconfiguration, plug-and-play and other quality of service functions. Multiple UAVs, flying in a swarm, constitute a network of distributed (in the spatio-temporal sense) sensors that must be coordinated to complete a complex mission. Current R&D activities are discussed that concern issues of modeling, planning and control. Here, optimum terrain coverage, target tracking and adversarial reasoning strategies require new technologies to deal with issues of system complexity, uncertainty management and computational efficiency [Vachtsevanos, et al, 2004]. We will pose the major technical challenges arising in the “system of systems” approach and state the need for new modeling, networking, communications and computing technologies that must be developed and validated if such complex unmanned systems as UAVs are to perform effectively and efficiently, in conjunction with manned systems, in a variety of application domains. We will conclude by proposing possible solutions to these challenges.