For decades, PID (Proportional + Integral + Derivative)-like controllers have been successfully used in academia and industry for many kinds of plants. This is thanks to its simplicity and suitable performance in linear or linearized plants, and under certain conditions, in nonlinear ones. A number of PID controller gains tuning approaches have been proposed in the literature in the last decades; most of them off-line techniques. However, in those cases wherein plants are subject to continuous parametric changes or external disturbances, online gains tuning is a desirable choice. This is the case of modular underwater ROVs (Remotely Operated Vehicles) where parameters (weight, buoyancy, added mass, among others) change according to the tool it is fitted with. In practice, some amount of time is dedicated to tune the PID gains of a ROV. Once the best set of gains has been achieved the ROV is ready to work. However, when the vehicle changes its tool or it is subject to ocean currents, its performance deteriorates since the fixed set of gains is no longer valid for the new conditions. Thus, an online PID gains tuning algorithm should be implemented to overcome this problem. In this paper, an auto-tune PID-like controller based on Neural Networks (NN) is proposed. The NN plays the role of automatically estimating the suitable set of PID gains that achieves stability of the system. The NN adjusts online the controller gains that attain the smaller position tracking error. Simulation results are given considering an underactuated 6 DOF (degrees of freedom) underwater ROV. Real time experiments on an underactuated mini ROV are conducted to show the effectiveness of the proposed scheme.

Underwater remotely operated vehicles (ROVs) play an important role in a number of shallow and deepwater missions for marine science, oil and gas extraction, exploration and salvage. In these applications, the motions of the ROV are guided either by a human pilot on a surface support vessel through an umbilical cord providing power and telemetry, or by an automatic pilot. In the case of automatic control, ROV state feedback is provided by acoustic and inertial sensors and this state information, along with a controller strategy, is used to perform several tasks such as station-keeping and autoimmersion/heading, among others. In this paper, the modelling, design and control of the Kaxan ROV is presented: i) The complete six degrees of freedom, non linear hydrodynamic model with its parameters, ii) the Kaxan hardware/software architecture, iii) numerical simulations in Matlab/Simulink platform of a model-free second order sliding mode control along with ocean currents as disturbances and thruster dynamics, iv) a virtual environment to visualize the motion of the Kaxan ROV and v) experimental results of a one degree of freedom underwater system.

PID controllers are widely used and adaptable to various types of systems. However, for the response to be adequate under different conditions, the PID gains must be adjusted. The tuning is made according to the difference between the reference value and the real value (error). This work presents a selfadjusting PID controller based on a backpropagation artificial neural network. The network calculates the appropriate gains according to the desired output, that is, the dynamic response desired which is composed of the transient part and the stationary part of the step response of a system. The contribution of the work is that in addition to using the error for network training, the maximum desired values of overshoots, settling times, and stationary errors were used as input data for the network. An offline training database was created using genetic algorithms to obtain the dynamic response data associated with PID gains. The genetic algorithm allows getting data in different operating ranges and allows using only stable gains combinations. The database was used for training. Subsequently, the neural network estimates an appropriate gain combination, adapting to the error and the desired response. The method performance is evaluated by controlling the speed of a direct current motor. The results indicate an average error of 4% for the database between the requested and system response. On the other hand, the gains estimated by the network in the test dataset (1544 combinations) did not cause instability and complying with the expected dynamic response in 86% of the dataset.

This paper presents the improvement of an ultrasonic pulse generator for a pipeline inspection gauge (PIG), which uses 64 transducers for inspecting distances up to 100 km with an axial resolution fixed at 3 mm and variable speeds between 0 and 2 m/s. An ultrasonic pulse generator is composed of a high-voltage (HV) MOSFETs, driver logic and an HV power supply. We used a DC-HV DC converter device as the HV power supply because it reduces the size of the ultrasound system considerably. However, pipeline geometry and inspection effects such as hammer and shock cause a variable pulse repetition frequency (PRF), producing voltage drops, poor quality of the HV pulse generated, failures in the dimensioning of defects and damage to devices by over-voltage. Our improvement is to implement a control scheme to maintain the high quality of the HV regardless of the variable PRF. To achieve this, we characterized three transfer functions of the DC-HV DC converter, varying the connected load to 10%, 45% and 80%. For the characterization, we used the least squares technique, considering an autoregressive exogenous (ARX) model. Later, we compared three control schemes: (1) proportional-integral-derivative (PID) tuned by simultaneous optimization of several responses (SOSR), (2) PID tuned by a neural network (NN) and (3) PI tuned by the analytical design method (ADM). The metrics used to compare the control schemes were the recovery time, the maximum over-voltage and the excess energy when the shock and hammer effects happen to occur. Finally, to verify the improvement of the HV pulser, we compared the ultrasonic pulses generated for various frequencies and amplitudes using the pulse generator with and without the control scheme.

Backstepping is a control technique based on Lyapunov’s theory that has been successfully implemented in the control of motors and robots by several nonlinear methods. However, there are no standardized methods for tuning control gains (unlike the PIDs). This paper shows the tuning gains of the backstepping controller, using Genetic Algorithms (GA), for an Unmanned Aerial Vehicle (UAV), quadrotor type, designed for autonomous trajectory tracking. First, a dynamic model of the vehicle is obtained through the Newton‒Euler methodology. Then, the control law is obtained, and self-tuning is performed, through which we can obtain suitable values of the gains in order to achieve the design requirements. In this work, the establishment time and maximum impulse are considered as such. The tuning and simulations of the system response were performed using the MATLAB-Simulink environment, obtaining as a result the compliance of the design parameters and the correct tracking of different trajectories. The results show that self-tuning by means of genetic algorithms satisfactorily adjusts for the gains of a backstepping controller applied to a quadrotor and allows for the implementation of a control system that responds appropriately to errors of different magnitude.