Direct Motion Planning for Vision-Based Control

Abstract

This paper presents direct methods for vision-based control for the application of industrial inkjet printing. In this, visual control is designed with a direct coupling between camera measurements and joint motion. Traditional visual servoing commonly has a slow visual update rate and needs an additional local joint controller to guarantee stability. By only using the product as reference and sampling with a high update rate, direct visual measurements are sufficient for controlled positioning. The proposed method is simpler and more reliable than standard motor encoders, despite the tight real-time constraints. This direct visual control method is experimentally verified with a 2D planar motion stage for micrometer positioning. To achieve accurate and fast motion, a balance is found between frame rate and image size. With a frame rate of 1600 fps and an image size of 160 100 pixels we show the effectiveness of the approach. Note to Practitioners-This method of visual control is motivated by the current state-of-the-art in display manufacturing. Traditional solutions for the manufacturing of displays assumes that the transformation between production head and control reference (e.g., kinematic or dynamic model) is known with high accuracy and that the display itself is rigid (i.e., fixed pitch between pixel centers). For flexible displays, this latter assumption is not the case. A method is proposed that takes these issues into account. A camera measures directly where the center of each display cell is located and generates online a trajectory for velocity motion control.This velocity reference is based on a smooth profile with a fixed velocity on cell centers and a higher velocity in between cell centers. This enables a higher overall velocity while ensuring a similar quality of printing compared to a constant velocity reference. Visual control and trajectory generation is executed at 1600 [Hz] with an image size of 160 100 pixels. Feedback is obtained only from visual input, the encoders present in the motors are not used.

3D Printing Technology Toward State‐Of‐The‐Art Photoelectric Devices

“…High resolution and multi‐layer printing was developed to fabricate light‐emitting and hole transport layers via inkjet printing. [ 105 ] Moreover, multi‐layer printing is achieved by controlling the wetting conditions test and visually aligning. Afterward, the surface treatment technique was adopted to make a precise pattern.…”

Cross-Domain Modeling and Optimization of High-Speed Visual Servo Systems

“…A wafer with pre-defined micro-patterns on it is mounted on the motion stage. This prototype is applied to a use case of visionbased inkjet printing on non-uniform micro-structures, as detailed in [19]. In such a use case, visual measurements are used in a closed loop to compensate for deformations of flexible display substrates.…”

Positioning and navigation of mobile robot with asynchronous fusion of binocular vision system and inertial navigation system

“…The gradient descent algorithm is adopted to calculate the attitude quaternion q r for the accelerometer and the magnetometer. According to the principle of gradient descent, the quaternion q r is given by q r;k ¼ q r;k À 1 À l rf ða b ; b n ; m b Þ jjrf ða b ; b n ; m b Þjj (11) where l is the step size and rf ða b ; b n ; m b Þ is the gradient of equation 10rf ða b ; b n ; m b Þ ¼ J T ðqÞf ða b ; b n ; m b Þ…”

“…Moreover, positioning accuracy achieved by this fusion method is better than that of traditional relative positioning technologies, such as wheeltype mileage meters, inertial navigators, radars, and sonar, meaning that it is an effective alternative to the traditional positioning methods. 11 The common form of vision-aided inertial navigation systems (INSs) combines a monocular or binocular camera system with a low-cost IMU, which has been widely used in both robotics and autonomous vehicle navigation. [12][13][14][15][16][17][18] Largely benefited by this method, positioning error caused by the uncertainty of visual feature extraction in the space can be reduced, the drift phenomenon can be diminished, and the accumulated error in the positioning of inertial measurement can be overcome.…”