Motion planning and posture control of multiple n-link doubly nonholonomic manipulators

Abstract

SUMMARYThe paper considers the problem of motion planning and posture control of multiple n-link doubly nonholonomic mobile manipulators in an obstacle-cluttered and bounded workspace. The workspace is constrained with the existence of an arbitrary number of fixed obstacles (disks, rods and curves), artificial obstacles and moving obstacles. The coordination of multiple n-link doubly nonholonomic mobile manipulators subjected to such constraints becomes therefore a challenging navigational and steering problem that few papers have considered in the past. Our approach to developing the controllers, which are novel decentralized nonlinear acceleration controllers, is based on a Lyapunov control scheme that is not only intuitively understandable but also allows simple but rigorous development of the controllers. Via the scheme, we showed that the avoidance of all types of obstacles was possible, that the manipulators could reach a neighborhood of their goal and that their final orientation approximated the desired orientation. Computer simulations illustrate these results.

A dϕ-Strategy: Facilitating Dual-Formation Control of a Virtually Connected Team

“…Swarms or flocks in nature are commonly dovetailed with formation types and unique patterns, seen, for example, in schools of fishes, flocks of birds, swarms of insects, and herds of animals. In addition, a wide spectrum of the "formation rigidity" has also appeared in literature using the nomenclature from [2,3]; on one end there are split/rejoin maneuvers also known as minimally rigid formations [2,3] which are required in applications such as reconnaissance, sampling, and surveillance, while on the other end there are globally rigid formations which are required in applications that require cooperative payload transportation [4][5][6]. Then there are locally rigid formations (required in convoying and demining [1,7]) which are not infinitesimally rigid and allow for slight distortions temporarily.…”

“…Formation control can be defined as controlling the configuration or state of a team of agents en route to a target, normally maintaining constant their relative locations, hence maintaining the formation type and pattern [8][9][10][11]. Indeed, scalable formations with increased heterogeneity are evidently deployed to satisfy stringent time, labor, and cost demands.…”

“…A number of approaches such as leader-follower, virtual structure, nearest neighbors, social potentials, behaviorbased, and formation-constrained functions have been utilized to address formation control [11]. While the leadcarrier approach is known for its poor disturbance rejection properties [15] and dependence placed upon a single agent, it is commonly utilized in literature because of its simplicity, scalability, and ability to contain an array of formations with richer specifications and complexities [3,9,11,[15][16][17].…”

“…While the leadcarrier approach is known for its poor disturbance rejection properties [15] and dependence placed upon a single agent, it is commonly utilized in literature because of its simplicity, scalability, and ability to contain an array of formations with richer specifications and complexities [3,9,11,[15][16][17]. This work deploys a new lead-carrier scheme to establish and control a team fixed in a dual-formation, for the first time.…”

“…Since then the literature has grown with research using Type ( ℎ, ℎ) mobile manipulators (see, e.g., [11,[19][20][21]). The challenges associated with motion planning and control of the Type ( ℎ, ℎ) are amplified by the intimate coupling of the nonholonomic and holonomic constraints arising from the amalgamation of an articulated robotic arm and a wheeled platform.…”