New advances in multiple autonomous aerial robots formation control technology

Yang Xu, DeLin Luo, YanCheng You, HaiBin Duan
2019 Science China Technological Sciences  
Autonomous aerial robotics has become a hot direction of research inside the community of robotics and control. The primary problem addressed by formation control is to steer multiple aerial robots to form desired geometric patterns and, at the same time, realize desired collective swarming behaviors in a decentralized or distributed manner. In contrast to ground vehicles, aerial robots have the ability to work in three-dimensional (3D) airspace. Equipped with electric or hydraulic motors, the
more » ... ertical take-off and landing (VTOL) capability is a typical performance of aerial robots. Formation control technology for such aerial robots is incessantly springing up to satisfy the requirements of highly intelligent autonomous systems, which affects both military and civil areas, including missile defense, battlefield surveillance, satellite network construction, fire suppression, power grid inspection, commercial show, etc. [1][2][3][4][5]. Such a problem of multiple aerial robots formation control is exceptionally challenging to analyze if practical constraints such as complex dynamics, motion constraints, and imperfect measurements are incorporated. The problem of formation control has been studied for several decades. The approaches proposed in the early stage such as behavior-based ones can handle complex formation tasks but are not able to guarantee system convergence. Ever since the successful application of the consensus theory in the formation control, tremendous research efforts have been devoted to developing convergence-guaranteed formation control approaches. Very recently, many new methods have been proposed to push the boundary of formation control technologies towards practical applications on aerial robots. This news aims to survey the most recent advances in the formation control, which can be directly or potentially applied to autonomous aerial robots. Based on consensus theory as well as algebraic graph theory, innovative formation control schemes have been proposed in recent surveys [6,7]. Most of graphical formation control methods try to encode certain algebraic constraints related to the target configurations with specific Laplacian matrices distributedly, such as complex Laplacian, stress matrix, signed Laplacian, barycentric coordinate, and rigidity theories. Complex Laplacian-based formation control is confined in the plane due to the definition of the complex coordinate. Formation rotating maneuvers can be achieved by the barycentric coordinate-based approach, but the corresponding control law is quite complicated by introducing the coordinate transform matrix. Stress matrixbased and rigidity theories-based formation methods require the interaction networks to be undirected, and thus resultant strict graphical conditions are not suitable for implementations. Notably, by combing properties of both affine transformation and signed Laplacian, signed Laplacian-based formation control could be applied to any dimensional coordinate-free condition under generally directed interaction networks [8]. Furthermore, equipped with different onboard sensors, researchers can categorize the formation control into the position-, displacement-, bearing-, and distance-based
doi:10.1007/s11431-018-9457-9 fatcat:r64vftwrh5cv5kqrntzru5vzj4