Hector M. Becerra

Combining fuzzy, PID and regulation control for an autonomous mini-helicopter

This paper reports on the synthesis of different flight controllers for an X-Cell mini-helicopter. They are developed on the basis of the most realistic mathematical model currently available. Two hybrid intelligent control systems, combining computational intelligence methodologies with other control techniques, are investigated. For both systems, Mamdani-type fuzzy controllers determine the set points for altitude/attitude control. These fuzzy controllers are designed using a simple rule base. The first scheme consists of conventional SISO PID controllers for z-position and roll, pitch and yaw angles. In the second scheme, two of the previous PID controllers are used for roll and pitch, and a linear regulator is added to control altitude and yaw angle. These control schemes mimic the action of an expert pilot. The designed controllers are tested via simulations. It is shown that the designed controllers exhibit good performance for hover flight and control positioning at slow speed.

A Sliding-Mode-Control Law for Mobile Robots Based on Epipolar Visual Servoing From Three Views

Driving mobile robots to precise locations is of recognized interest, and using vision sensors in this context supplies many advantages. We propose a novel control law based on sliding-mode theory in order to drive mobile robots to a target location, which is specified by a previously acquired reference image. The control scheme exploits the piecewise epipolar geometry of three views on the basis of image-based visual servoing, in such a way that no 3-D scene information is required. The contribution of the paper is a new control law that achieves convergence to the target with no auxiliary images and without changing to any approach other than epipolar-based control. Additionally, the use of sliding-mode control deals with singularities, thus allowing the robot to move directly toward the target as well as avoiding the need of a precise camera calibration. The effectiveness of our approach is tested with simulations and real-world experiments.

Omnidirectional visual control of mobile robots based on the 1D trifocal tensor

The precise positioning of robotic systems is of great interest particularly in mobile robots. In this context, the use of omnidirectional vision provides many advantages thanks to its wide field of view. This paper presents an image-based visual control to drive a mobile robot to a desired location, which is specified by a target image previously acquired. It exploits the properties of omnidirectional images to preserve the bearing information by using a 1D trifocal tensor. The main contribution of the paper is that the elements of the tensor are introduced directly in the control law and neither any a priori knowledge of the scene nor any auxiliary image are required. Our approach can be applied with any visual sensor obeying approximately a central projection model, presents good robustness to image noise, and avoids the problem of a short baseline by exploiting the information of three views. A sliding mode control law in a square system ensures stability and robustness for the closed loop. The good performance of the control system is proven via simulations and real world experiments with a hypercatadioptric imaging system.

Combining fuzzy and PID control for an unmanned helicopter

This paper reports the synthesis of a controller for the X-cell minihelicopter. It is developed on basis of the most realistic mathematical model actually made available by V. Gabrilets et al. A combined control structure is proposed: Mamdani controllers keep set points for an altitude/attitude controller. These controllers are designed in the simplest rule base. Altitude/attitude controller is constituted for conventional SISO PID controllers for z-position and roll, pitch and yaw angles. This control scheme mimics the action of an expert pilot. The proposed scheme is tested via simulations; it presents a good performance for hover flight, and control position in slow speed.

A Sliding Mode Control law for epipolar visual servoing of differential-drive robots

In this paper, a robust control technique (sliding mode control) is proposed to be used in order to perform visual servoing for differential-drive mobile robots using the classical teach-by-showing strategy. We propose a commuted sliding mode control law that exploits the epipolar geometry. The major contribution of the paper is the design of a control law that solves the problem of passing through a singularity induced by the epipoles maintaining bounded inputs. Moreover, the designed control is able to drive the robot to the target even when it just starts on the singularity. The proposed approach does not need a precise camera calibration due to the robustness of the control system under uncertainty in parameters. It also ensures entire correction of both orientation and lateral error even with noise in the image. The effectiveness of our approach is tested via simulations.

Wheeled mobile robots navigation from a visual memory using wide field of view cameras

In this paper, we propose a visual path following control scheme for wheeled mobile robots based on the epipolar geometry. The control law only requires the position of the epipole computed between the current and target views along the sequence of a visual memory. The proposed approach has two main advantages: explicit pose parameters decomposition is not required and the rotational velocity is smooth or eventually piece-wise constant avoiding discontinuities that generally appear when the target image changes. The translational velocity is adapted as required for the path and the approach is independent of this velocity. Furthermore, our approach is valid for all cameras obeying the unified model, including conventional, central catadioptric and some fisheye cameras. Simulations as well as real-world experiments with a robot illustrate the validity of our approach.

Pose-estimation-based visual servoing for differential-drive robots using the 1D trifocal tensor

A pose-estimation-based approach to perform visual control for differential-drive robots is presented in this paper. Our scheme recovers the camera location (position and orientation) using an Extended Kalman Filter (EKF) algorithm with the 1D trifocal tensor (TT) as measurement model. This new visual servoing scheme allows knowing the real world path performed by the robot without the computational load introduced by position-based approaches. A state-estimated feedback control law is designed to solve a tracking problem for the lateral and longitudinal robot coordinates. The desired trajectories to be tracked ensure total correction of both position and orientation using a single control law, even though the orientation is a DOF in the control system. The effectiveness of our approach is tested via simulations.

Exploration of an unknown environment with a differential drive disc robot

This paper addresses the problem of exploring an unknown, planar, polygonal and simply connected environment. A saliency object (i.e. a landmark) is located in the environment. The collision-free subset of the robots configuration space is simply connected or it might have several connected components. The robot is a differential drive system shaped as a disc. The robot has limited sensing, namely it is incapable of measuring any distance or angle, or performing self localization. The exploration problem consists in discovering the environment with the robots sensor. To solve this problem, a motion policy is developed based on simple sensor feedback and a complete exploration strategy is represented as a Moore Machine. The proposed exploration strategy guarantees that the robot will discover the largest possible region of the environment. Consequently, the robot will find the landmark or declare that an exploration strategy to find it does not exist.

Exploiting the Trifocal Tensor in Dynamic Pose Estimation for Visual Control

Image-based approaches for visual control are memoryless, and they depend on the information extracted from the image plane. We propose the use of dynamic pose estimation in the task of driving a mobile robot to a desired location specified by a target image. This approach reduces the dependence of the control on the quality of current visual data and facilitates the planning of complex tasks. The pose estimation exploits the 1-D trifocal tensor (TT) as measurement, which allows us to obtain a semicalibrated estimation scheme that is valid for any visual sensor obeying a central projection model. The contribution of this brief is a novel observability analysis of the estimation problem from the 1-D TT using nonlinear tools, as well as the demonstration of the validity of closed-loop control from the estimated pose by showing a separation principle in our nonlinear framework. The overall position-based scheme drives the robot to a desired pose through smooth velocities without the need of a target model, either scene reconstruction or depth information. The effectiveness of the approach is evaluated via real-world experiments.

A novel 1D trifocal tensor-based control for differential-drive robots

This paper presents an image-based approach to perform visual control for differential-drive robots. We use for the first time the elements of the 1D trifocal tensor directly in the control law. The visual control utilizes the usual teach-by showing strategy without requiring any a prior knowledge of the scene and does not need any auxiliary image. The main contribution of the paper is that the proposed two-steps control law ensures total correction of both position and orientation without switching to any other visual constraint rather than the 1D trifocal tensor. The paper exploits the sliding mode control technique in a square system, ensuring stability and robustness for the closed loop. The good performance of the control system is proven via simulations.