Tunnel passing maneuvers of prescribed formations

Abstract

SUMMARYTunnel passing is a pattern formation of multiple robots, an outcome of formation control which is the general problem of controlling a large number of robots required to move as a group. Tunnel passing deals with the task of driving a team of robots from arbitrary initial positions through a tunnel of given geometry. This paper proposes a decentralized planner that guarantees collision-free tunnel passing maneuvers of a team of nonholonomic car-like robots fixed in a prescribed formation, while considering all the practical limitations and constraints due to nonholonomy, tunnel geometry, and the formation specifications. Although solutions in literature are restricted to tunnels with linear segments, this paper introduces piecewise tunnel walls with straight and curved segments. The motion planner, derived from the Lyapunov-based control scheme works within an overarching leader-follower framework to generate either split/rejoin or expansion/contraction of the formation, as feasible solutions. Results from simulating virtual scenarios demonstrated the effectiveness of the proposed nonlinear controllers.

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

“…This is where all the carrier robots are virtually connected to the lead robot along the trajectory to task completion. The concept inspired by the work by Consolini et al in 2008 in [15] and Sharma et al in 2012 in [25], is classified as thestrategy where and are Euclidean measures of a leadcarrier robot scheme. Fixing the values of enables us to maintain a fixed distance between the lead robot and each of its carriers.…”

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

“…Adopting the nomenclature and methodology of Sharma et al in, 27 we assume that thelth curveshaped obstacle can be collapsed into an arc in the z 1 -z 2 plane with initial coordinates (m˜l 1 , n˜l 1 ) and final coordinates (m˜l 2 , n˜l 2 ), which can be extended from a center (ca˜l, cb˜l) (see Fig. 5).…”

A new stabilizing solution for motion planning and control of multiple robots

“…The workspace is a fixed, closed and bounded rectangular region for some η 1 > 0 and η 2 > 0. Precisely, the workspace is the set W S = {(z 1 , z 2 …”

“…Thus if (x i − a j cos θ j + (y i − b j ) sin θ j < 0, then we take λ (2) ij = 0 and if (x i − a j cos θ j + (y i − b j ) sin θ j > l j + r j , then we take λ (2) ij = l j + r j . For i, j = 1, 2 and i = j , we further define…”

“…1,[3][4][5] Developing multiple robots are favored over single units due to the stringent time and cost constraints, increased robustness, greater fault tolerance, better safety, accelerated performances and higher capabilities, to outline a few major ones. 2,3,6 The multiple robots can cooperate and network for better, faster and more efficient results.…”