Research progress for cooperative control of heterogeneous unmanned systems

Recent advancements in artificial intelligence and control theory have significantly enhanced unmanned systems, endowing them with remarkable capabilities, such as autonomy, self-learning, and scalability. These innovations are pushing unmanned systems toward more distributed and collaborative operational modes. However, the requirements for complex, diverse and stereoscopic missions pose great challenges to unmanned systems. To address such issues, cross-domain collaboration across air, land, and maritime environments using heterogeneous unmanned systems (HUS) offers a promising solution. Such collaboration not only boosts individual system capabilities but also builds a technological foundation for future cooperative operations and smart living scenarios. This paper provides a comprehensive review of the literature on cross-domain collaboration from both domestic and international perspectives. It also discusses the practical engineering applications of HUS in military, scientific research, and civilian fields. Currently, technological progress in this field has been notably spurred by military powers, particularly the United States, the United Kingdom, and France. Although China’s development is still in its nascent stages, it has made significant strides in theoretical and technological research. Research into cross-domain collaboration of HUS involves sophisticated theoretical frameworks and interdisciplinary integration. This paper elaborates on the latest research developments in collaborative control of HUS, focusing on four aspects: consensus, trajectory tracking, formation-containment, and diverse communication scenarios. Given the difference between the theoretical aspects of upper-level control and the feasibility of lower-level control, research on trajectory tracking methods is divided into four main types: sliding mode control, adaptive control, backstepping control, and adaptive dynamic programming. Each control method has its unique strengths, limitations, and suitable scenarios. Despite the rich body of research on cooperative control of HUS, numerous challenges remain in dealing with unknown and complex environments. This paper summarizes previous relevant work and underscores technical challenges across three primary areas: the conflict between multiple constraints and real-time task performance, the issue of multitask switching control, and the problem of maintaining control stability problem under cross-domain communication. Concrete examples are provided to elucidate the difficulties in achieving effective collaborative control for HUS. Based on the current research status and practical requirements for collaborative control of HUS, future development will likely focus on three key areas: deep reinforcement learning and distributed games, human–machine interaction, and antiswarm intelligence. In summary, the broad applicability and strategic importance of collaborative control of HUS have attracted attention from leading military powers, solidifying its status as a pivotal area of technological innovation.