Visual servoing of mobile robots for posture stabilization: from theory to experiments

Abstract

On the basis of the kinematic model of a unicycle mobile robot in polar coordinates, an adaptive visual servoing strategy is proposed to regulate the mobile robot to its desired pose. By regarding the unknown depth as model uncertainty, the system error vector can be chosen as measurable signals that are reconstructed by a motion estimation technique. Then, an adaptive controller is carefully designed along with a parameter updating mechanism to compensate for the unknown depth information online. On the basis of Lyapunov techniques and LaSalle's invariance principle, rigorous stability analysis is conducted. Because the control law is elegantly designed on the basis of the polar-coordinate-based representation of error dynamics, the consequent maneuver behavior is natural, and the resulting path is short. Experimental results are provided to verify the performance of the proposed approach.The vision-based robotic system is composed of a mobile robot and an onboard camera fixed on a pan motion platform. As shown in Figure 1, the robot frame F R is attached to the center of the wheel axis, which is defined as the Y R axis, while the Z R axis is perpendicular to the motion plane of the mobile robot, and the X R axis can be then obtained easily for a standard right-handed coordinate system. The camera frame F C is defined such that the camera center O C lies in the Z R axis and the frames F R and F C are parallel to each other when the rotation angle of the pan camera is zero. The From the previous analysis, it is clear that the largest invariant set M only consists of the equilibrium point in the sense that‡ It should be noted that although P O c 0 .t/ is, as can be seen from (15), merely piecewise smooth because of the usage of the projection function, it is however continuous with respect to initial conditions. Then, on the basis of Lemma 1 in [32], we can conclude that the positive limit L C of P O c 0 .t/ is invariant, which thus makes LaSalle's invariance principle applicable to the closed-loop error system of (20).

Vision-Based Model Predictive Control for Steering of a Nonholonomic Mobile Robot

“…5N u + 6N) 3 ) time. Therefore, the total time cost of the QUADPROG method is O(N 4 + N + (4Nu + 6N) × Nu 2 + (5Nu + 6N) 3 ). Consider the dynamics of the i th neuron of the LVI-PDNN in (57) employed to solve QP (31)- (33).…”

“…V ISUAL feedback-based motion control of nonholonomic mobile robots (NMRs) has been a popular topic in robotics research in recent decades [1]- [3], [8]. Most studies on position-based visual servoing carried out in this field involve reconstruction of the target pose with respect to the robot, and thus leads to a Cartesian motion control problems, such as vision-based tracking and vision-based stabilization.…”

Model-free unified tracking and regulation visual servoing of nonholonomic mobile robots

“…As a unified controller can be utilized for both tracking and regulation problems, it avoids the frequent switch between different control laws, and brings much convenience for practical systems. On the other hand, a significant issue with vision system is the lack of depth information, which prevents robot pose construction from vision signals without a priori information about target object [6] [7]. Therefore, it is very challenging, and equally important to solve the unified tracking and regulation visual servoing problem for mobile robots.…”

Trajectory Tracking of an Omni-Directional Wheeled Mobile Robot Using a Model Predictive Control Strategy

“…In recent decades, due to the broad applications for mobile robots, the core issues in motion control of mobile robots, including point stabilization, path planning, trajectory tracking, and real-time avoidance, have attracted considerable attention from both academic scholars and practitioners [1][2][3][4][5].…”

“…The optimal solution is taken as the optimal control input in the future prediction time domain and the first control vector of the optimal solution is used as the real control input. The advantages of the MPC algorithm are: (1) It is easy to model; (2) It has a rolling optimization strategy with good dynamic control effect; (3) It can correct the output by feedback which improves the robustness of the control system; (4) As a computer optimization control algorithm, it is easy to realize on a computer. In recent years, the MPC algorithm has also been widely used to achieve optimized motion control of mobile robots.…”