CAD Model-based Tracking and 3D Visual-based Control for MEMS Microassembly

Abstract

This paper investigates sequential robotic microassembly for the construction of 3D micro-electro-mechanical systems (MEMS) structures using a 3D visual servoing approach. The previous solutions proposed in the literature for these kinds of problems are based on 2D visual control because of the lack of precise and robust 3D measures from the work scene. In this paper, the relevance of the real-time 3D visual tracking method and the 3D vision-based control law proposed is demonstrated. The 3D poses of the MEMS are supplied in real-time by a computer-aided design (CAD) model-based tracking algorithm. This latter is sufficiently accurate and robust to enable a precise regulation toward zero of the 3D error using the proposed pose-based visual servoing approach.Experiments on a microrobotic setup have been carried out to achieve assemblies of two or more 400 µm × 400 µm × 100 µm silicon micro-objects by their respective 97 µm × 97 µm × 100 µm notches with an assembly clearance from 1 µm to 5 µm. The different microassembly processes are performed with a mean error of 0.3 µm in position and 0.35×10 −2 rad in orientation.

Dynamic grasp and trajectory planning for moving objects

“…Nevertheless, it is worth mentioning that due to the highly modular nature of our implementation, the current pose tracker can be easily replaced with another one e.g. a CAD model-based tracker (Ortenzi et al 2016;Tamadazte et al 2010) or a coupled layer pose tracker (Ma et al 2016).…”

Highly precise micropositioning task using a direct visual servoing scheme

“…In the last decade considerable researches have been performed on the development of robotic microassembly station, gripping systems, precise actuators, and microassembly strategies. Many works have dealt with the development of control approaches to automate the different micromanipulation and microassembly tasks such positioning, orientation, picking, placing, and insertion of the different micro-objects [10], [8]. Most of these works considered the use of the vision system to control the behavior of the robotic structure of the microassembly station during the assembly process.…”

“…Most of these works considered the use of the vision system to control the behavior of the robotic structure of the microassembly station during the assembly process. Thus, the control of robots through realtime and continuous visual feedback is generally known as visual servoing [4], and the continuous observation of the objects of interest is referred to visual tracking [11], [10]. Visual tracking of an object involves the detection of some known object features in the acquired images and, using these features, it is possible to estimate the the object position and orientation.…”

“…Considering that the control of robot is directly linked to the variation of the pixels intensity in the image, no geometric information has to be extracted and, consequently, tracked or matched over time. This avoid the development of complex visual tracking algorithms such as in [10], [11]. Let us also advocate that avoiding this feature extraction process allows to be naturally robust to error measurements increasing accuracy and repeatability of the positioning tasks.…”

Navigation of a rolling microrobot in cluttered environments for automated crystal harvesting

“…Micromanipulation has been proposed for such applications as micro-machining [4], cell manipulation [1], and MEMS microassembly [5]. In [6], [7], a rolling microrobot design called the RodBot was introduced for use in x-ray crystallography X-ray crystallography concerns the study of properties of biological molecules and proteins, which are crystallized and analyzed by small-angle x-ray scattering, and is used extensively in the pharmaceutical industry [8].…”