Control structure for a car-like robot using artificial neural networks and genetic algorithms

Flórez, Camilo Andrés Cáceres; Rosário, João Maurício; Amaya, Darío

doi:10.1007/s00521-018-3514-1

Cited by 16 publications

(2 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The combination of learning artificial neural networks using hybrid approaches (GA) is very current and is based on many studies. If these techniques are combined well, we can achieve better results than if done separately [21][22][23][24][25][26]. At the present time, methods of artificial neural networks and genetic algorithms are used in all possible scientific fields, from medicine to earth sciences.…”

Section: Related Workmentioning

confidence: 99%

The Influence of Genetic Algorithms on Learning Possibilities of Artificial Neural Networks

et al. 2022

View full text Add to dashboard Cite

The presented research study focuses on demonstrating the learning ability of a neural network using a genetic algorithm and finding the most suitable neural network topology for solving a demonstration problem. The network topology is significantly dependent on the level of generalization. More robust topology of a neural network is usually more suitable for particular details in the training set and it loses the ability to abstract general information. Therefore, we often design the network topology by taking into the account the required generalization, rather than the aspect of theoretical calculations. The next part of the article presents research whether a modification of the parameters of the genetic algorithm can achieve optimization and acceleration of the neural network learning process. The function of the neural network and its learning by using the genetic algorithm is demonstrated in a program for solving a computer game. The research focuses mainly on the assessment of the influence of changes in neural networks’ topology and changes in parameters in genetic algorithm on the achieved results and speed of neural network training. The achieved results are statistically presented and compared depending on the network topology and changes in the learning algorithm.

show abstract

Section: Related Workmentioning

confidence: 99%

The Influence of Genetic Algorithms on Learning Possibilities of Artificial Neural Networks

et al. 2022

View full text Add to dashboard Cite

show abstract

“…Narcis et al 14 proposed a solution for planning and controlling a car-like robot in environments cluttered has obstacles with low computational cost. Camilo et al 15 propose a control strategy for car-like robots, specifically developing the obstacles avoidance and position control to reach the desired position of the robot in an unknown environment by integrating potential fields with neuroevolution controllers and trained using a designed training environment. Zhang et al improved the A* hybrid algorithm to find a feasible path plan of the spherical robot with smoothing curves and adding speed limit to the entire path tight bends and small.…”

Section: Introductionmentioning

confidence: 99%

Path tracking control for car-like robots by pid controller with time-varying parameters

Trinh

Nguyêñ

2022

Sci.Tech.Dev.J.-Eng.Tech.

View full text Add to dashboard Cite

Wheel mobile robots have been widely applied to many areas of life, from serving people in services and industrial logistics to military fields. Applications of wheeled mobile robots such as floor cleaning robots, supermarket robots, logistics robots in hospitals, autonomous robots that guide automatic guidance in intelligent factories, military robots and agricultural robots. Thereby, we can see the great potential of mobile robots in the future. However, among the wheel mobile robots, the differential-drive mobile robot structure is most commonly used because of its simplicity in structure and control. Nevertheless, the navigation and driving are done synchronously through the two drive wheels, which leads to slippage and difficulty in controlling posture errors when navigating, especially on roads curved with a small radius. In order to overcome the disadvantages of the steering and navigation of differential-drive mobile robots, the structure of the automobile has been applied to design a wheel-mobile robot called a car-like robot. Many studies have shown that the advantage of this design option is that the navigation control and robot drive are separated. As a result, the dynamic control loop can ignore the navigation part, while the kinematic model determines the navigation law. Therefore, the work proposes the design of a PID controller with time-varying parameters for a car-like robot to track a given trajectory with minimum error. The nonlinear kinematic model of the robot is linearized along a reference trajectory, and the obtained linear model is used in the controller design process. A PID controller is designed where the controller parameters are tuned to minimize the tracking error. Our model achieves a tracking error value, including the minor position is 5.1 mm, and the maximum postural error is 2o. The simulation results showed the proposed controller's effectiveness in position and posture errors, and it shows we can apply this research to control car-like mobile robots in logistics services.

show abstract

Proposal of a Smart Hospital Based on Internet of Things (IoT) Concept

Cáceres

Rosário

Amaya

2019

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Control structure for a car-like robot using artificial neural networks and genetic algorithms

Cited by 16 publications

References 20 publications

The Influence of Genetic Algorithms on Learning Possibilities of Artificial Neural Networks

The Influence of Genetic Algorithms on Learning Possibilities of Artificial Neural Networks

Path tracking control for car-like robots by pid controller with time-varying parameters

Proposal of a Smart Hospital Based on Internet of Things (IoT) Concept

Contact Info

Product

Resources

About