Task‐Space Bearing‐Only Formation Control for Networked Robotic Manipulators Without Velocity Measurement

ABSTRACT In this article, two novel bearing‐only formation control schemes based on the passivity property are proposed for networked robotic manipulators, which are capable of achieving end‐effector formation utilizing onboard vision‐based sensors. In contrast to the existing methods, the developed strategy exhibits the following features. Firstly, it eliminates the requirement for velocity measurements and communication among manipulators, which improves the flexibility and maneuverability of the system while significantly reducing computational overhead. Secondly, we establish the exponential stability of the desired formation and possess the duality property if the manipulator is non‐redundant. Furthermore, by incorporating an approximate differentiation filter to compensate for unavailable velocity measurements, the measurement conditions for formation are further reduced, making the approach applicable to both redundant and non‐redundant manipulators. This modification enables manufacturers to eliminate the need for an additional sensor on each manipulator, thereby making the formation system more cost‐effective, reducing load, and facilitating implementation. Two simulations involving a group of two‐link robotic manipulators are conducted to validate the efficiency of the theoretical results.