Mohammed Abdel-Rahman MAREY

摘要: Many researches in robotics aim at the involvement of exteroceptive additional sensors which is strongly contributing to the ever growing demand to improve autonomy, robustness, flexibility and precision. Nevertheless good sensing abilities are essential to gain a higher flexibility and autonomy of robots. When robots operate in unstructured environments, the sensory information is included in the control loop. Many different sensors have been developed over the past years to fit the requirements of different but very specific tasks. Measurements gained by employing these sensors, that reflect variation in some phenomena, are used in the control loop to adapt the task execution to these variations. The most common sensors in robotics are force/torque sensors, laser range finders, ultrasonic sensors and camera sensors. Among all sensors, vision is the most complementary one with respect to the sensory information it provides to be included in the control loop since visual sensors are powerful means for a robot to perceive its environment. In particular, the use of visual feedback from sensor camera, when it is used in a correct manner, guarantees accurate positioning, robustness to calibration uncertainties, and reactivity to environmental changes.Visual servoing is a viable method for robotic control based on the utilization of visual information extracted from images to close the robot control loop. Various areas are involved in visual servoing as shown in Fig. 1.1. Visual information obtained from the image processing can be used to extracting 2D features. It can also be used to estimating pose parameters by employing a pose estimation algorithm …