Path Planning for Visual Servoing and Navigation Using Convex Optimization

Abstract

Exploiting visual cues to control systems for robotic applications is a promising idea. Practically, they are usable only if the end result accedes to the restrictions of the specific environment. These restrictions are like the limited field of view of the camera and physical constraints of the workspace of the robot. Hence, there is a need for a general framework that can be used adaptively across various environments. We develop such an algorithmic framework that is flexible to accommodate various kinds of constraints and generate a solution that is optimal in the sense of the considered error measure. We perform a constrained optimization on the error in a convex domain considering all the necessary constraints using convex optimization techniques and further extend it to nonconvex domains. We utilize branch-and-bound algorithm to divide the problem of optimizing over a range of rotations into simpler problems and solve for the optimal rotation. We demonstrate the performance of the algorithm by generating control signals in a simulated framework for visual servoing and in a real-world for robot navigation.

An Innovative Collision-Free Image-Based Visual Servoing Method for Mobile Robot Navigation Based on the Path Planning in the Image Plan

“…The outcomes of these two projects are used in a remote operation. The works presented in [ 24 , 25 , 26 , 27 , 28 , 29 ] seek to find a feasible trajectory in Cartesian space (3D) while avoiding obstacles and guaranteeing the visibility of the object in the image plane. An estimation of the pose of the robot and a priori knowledge of the obstacles are necessary.…”

An Accurate Path Planning Algorithm Based on Triangular Meshes in Robotic Fibre Placement

“…Path planning is involved in many applications, and there are lots of the methods to plan the desired paths [5]- [6]. Meanwhile, path planning is also one of the key technologies in RFP and it determines the distribution way of the fibre tows placed on the mould surface.…”

Cooperative and interaction control for underwater robotic vehicles

“…Alternatively, other researchers have adopted optimization techniques that aim at finding an optimal path with respect to various metrics such as the distance from image boundary, the length of the path and the energy [97]. In particular, several approaches based on Part I -Introduction and Problem Statement 10 convex optimization techniques [98][99][100] and the perceptual control manifold [101] have been presented. On the other hand, employing the Lagrange multipliers approach, the optimization algorithm proposed in [102] aims at finding optimal image feature paths with respect to visibility constraints as well as various robot motion constraints.…”