In this paper, we utilized two vision-based go-to-goal robot control approaches and compare them to the non-holonomic indoor mobile robot. In the proposed methods, an overhead camera image data is used as the input parameters when the robot wheel speeds are determined. The main purpose of this system is to minimize the complexity of conventional robot control kinematics and to provide an efficient control approach when controlling the wheel speeds and the robot orientation angle of the mobile robot. In addition to reducing the complexity of robot control kinematics, it is also intended to reduce systematic and non-systematic errors. Our proposed method is divided into three phases; the first phase is that the calibration of the overhead camera and working environment. At this phase, target and labels placed on the robot were detected and robot position information was extracted. The second phase is tracking the robot motion and extracts the control inputs’ such as position and orientation based on visual feature information. The third phase is controlled loop by using Graph and Angle based control approaches merged with Gaussian and Decision tree algorithms. We have briefly described these control approaches as follows. Graph-based decision tree control as (GDTC), Graph-based Gaussian control as (GGC), Angle-based decision tree as (ADTC) and Angle based Gaussian control as (AGC). Experiments have been performed for the non-holonomic mobile robot with the eye-out-device camera configuration in real-time. Our experimental results are presented to emphasize the efficiency of the methods in the presence of kinematic disturbances and without using any extra sensors, only overhead camera.
Discussion(0)
No comments yet. Be the first to comment.