3D Motion Planning for Image-Based Visual Servoing Tasks

Visual predictive control architecture based on image moments for manipulator robots

“…Beginning at the current discrete time k with the current image I k having the point features f p (k), a reference trajectory is designed from the image sequences I k+i , i = 1, hp with point features w p (k + 1|k) so that to obtain w p (k + hp|k) = f * p . The notation w p (k + i|k) specifies that the reference trajectory depends on the initial conditions f p (k) at discrete time k. Among the methods to generate the image plane trajectories for tracked points in an eye-in-hand system, we have chosen the 3D motion planning approach for image-based visual servoing task from [13]. Considering that I k is the initial image and f p (k) its' point features, I k+hp the final image with f * p the point features and a fixed object described by four point features assumed to be coplanar but not collinear.…”

“…If the angle axis representation u, θ khp is considered and θ d (q) = θ khp q [13], then R d (q) can be expressed by:…”

“…As in [13] a helicoidal reference trajectory was considered. Starting from the initial conditions f p (k), the reference trajectory is generated:…”

Self-calibrated visual servoing with respect to axial-symmetric 3D objects

“…Helicoidal trajectories exhibit the nice property of combining translations and rotations in a natural way. In [2], the special case of helicoidal planning with constant radius and axis identical to the instantaneous axis of motion has been discussed. In the present work, the general case of variable radius and arbitrary axis is addressed.…”

Calibration Free Image Point Path Planning Simultaneously Ensuring Visibility and Controlling Camera Path

“…Based on the partial pose reconstruction (hence modelfree), more recent work [13], [14], [15] addresses also the visibility issue, essentially by proposing some circular camera path, centred at the object. Yet all these controllers need some further manual adjustment to determine the curvature of the path.…”