Intelligent neuro-controller for navigation of mobile robot

Neural network based systems have been used in past years for robot navigation applications because of their ability to learn human expertise and to utilize this knowledge to develop autonomous navigation strategies. In this paper, neural based systems are developed for mobile robot reactive navigation. The proposed systems transform sensors' input to yield wheel velocities. Novel algorithm is proposed for optimal training of neural network. With a view to ascertain the efficacy of proposed system; developed neural system's performance is compared to other neural and fuzzy based approaches. Simulation results show effectiveness of proposed system in all kind of obstacle environments.

IFAC Proceedings Volumes, 1998

This paper describes a neural network model for the reactive navigation of a mobile robot. The system defines a series of reactive behaviors: stop, avoid , stroll , wall following, etc. depending on the information obtained from a set of proximity sensors distributed in the periphery of the robot. Reinforcing learning permits the adaptative navigation of the robot.

Since last decade, navigation of mobile robot has received considerable attention in the field of research due to its applications, where the involvement of human directly is difficult or dangerous. In this paper, we present the development of robust control algorithm for navigation of mobile robot using artificial intelligent technique (AIT). The ability of learning for neural network (an AIT) is used to develop a strong adaptive back stepping controller that does not requires the knowledge of the robot dynamic. Therefore, we combined the ANN and kinematic control technique to solve the localization problem for safe and smooth navigation at narrow corridors. To navigate robustly inside environment and reach the goal position safely; different types of sensors has been mounted on the robot body. ANN has been used to train the system for working environment. A Matlab simulation platform is used to validate the experimental result. Finally, the success of proposed control algorithm is verified through the simulation experiment, which shows its superior performance and disturbance rejection.

IEEE 2002 28th Annual Conference of the Industrial Electronics Society. IECON 02, 2002

This article presents an artificial neural network (ANN) structure applied to control a mobile robot movement in dynamically changing environments (environments wirh mobile obstacles), The proposed structure is a backward neural one. So, ir is based on past andfihrre positions, and on a optimal pre-established parh. The past pasirions provide rhe ANN with memory of the mobile robot previous positions. On the other hand, rhe future positions provide rhe ANN with a goal, i.e., where the robot shouldgo. Basedon this information, the robot da nor lose ifs goal, even (f if has to avoid an obstacle. The results show the eflciency ofthe ANXin aform ofsimulations

2012

A new paradigm of intelligent navigation system for mobile robot has been enriched with some common features like: criteria for optimal performance and ways to optimize design, structure and control of robot. With the growing need for the deployment of intelligent and highly autonomous systems, it would be beneficial to flawlessly combine robust learning capabilities of artificial neural networks with a high level of knowledge interpretability provided by fuzzy-logic. Fuzzy-neural network is able to build comprehensive knowledge bases considering sensor-rich system with real time constraints by adaptive learning, rule extraction and insertion, and neural/fuzzy reasoning. This technique is simulated and also compared with other simulation studies by previous researcher . The training for back propagation algorithm and its navigational performances analysis has been done in real experimental setup. As experimental result matches well with the simulation result, the realism of method i...

2014

Over recent years, automated mobile robots play a crucial role in various navigation operations. For any mobile device, the capacity to explore in its surroundings is essential. Evading hazardous circumstances, for example, crashes and risky conditions (temperature, radiation, presentation to climate, and so on.) comes in the first place, yet in the event that the robot has a reason that identifies with particular places in its surroundings, it must discover those spots. There is an increment in examination here due to the requisition of mobile robots in a solving issues like investigating natural landscape and assets, transportation tasks, surveillance, or cleaning. We require great moving competencies and a well exactness for moving in a specified track in these requisitions. Notwithstanding, control of these navigation bots get to be exceptionally troublesome because of the exceedingly unsystematic and dynamic aspects of the surrounding world. The intelligent reply to this issue ...

A combination of three separate neural network modules is employed to deal with the mobile robot control problem, giving very promising results. At the lower level, a neuro-fuzzy architecture, trained by reinforcement learning, steers the robot such that to avoid obstacles, exploiting ultrasonic sensor readings, and head to a target, given the heading error. ln the second and most important level, a topologically ordered Hopfield neural network performs global path finding in real time, using an environment map arranged on the fly. The third level involves an ordinary Hopfield neural network, used as an associative memory, which tries to match and complete the currently observed environment with one of the stored ones. Computer simulation validates the efficiency of the approach and shows its potential benefits.

2015

Autonomous robot is a type of the robot which can perform operation accurately in any environment. Currently, the demands of robots are increasing in the industry; the autonomous robots suffer from the path planning and obstacle avoidance. To overcome this, many approaches have been proposed but due the noisy environment, it is difficult for the robots to perform navigation. In this work, we proposed an efficient model for the robot navigation using neural networks and to overcome the path planning, kinematics based model is presented. By utilizing the kinematics based stability model, we are able to achieve the stability condition of the robot and it is used for the controlling the speed of the robot. Neural network is used for the obstacle prediction and avoidance. The simulation result presented that the proposed model is robust to noisy (hindrance) environment and it is able to avoid the obstacle precisely at the maximum speed of 0.02 to 1.0 m/s.

This paper presents a novel reactive navigation algorithm for wheeled mobile robots under non-holonomic constraints and in unknown environments. Two techniques are proposed: a geometrical based technique and a neural network based technique. The mobile robot travels to a pre-defined goal position safely and efficiently without any prior map of the environment by modulating its steering angle and turning radius. The dimensions and shape of the robot are incorporated to determine the set of all possible collision-free steering angles. The algorithm then selects the best steering angle candidate. In the geometrical navigation technique, a safe turning radius is computed based on an equation derived from the geometry of the problem. On the other hand, the neural-based technique aims to generate an optimized trajectory by using a user-defined objective function which minimizes the traveled distance to the goal position while avoiding obstacles. The experimental results demonstrate that the algorithms are capable of driving the robot safely across a variety of indoor environments.

1995

rlcccntly it. has been introduced a neural controller for a mobile rohot.llm!.learns both forward aml inverse odomctry of a dlffercnt.ial-drivc robot through an unsupervised learning-bydoing cycle. This article introduces an obst.aclc avoidance module that is integrated into the neural controller. This module makes use of sensory informat.ion to determine at each instant. a cksircd cmgle and distance that causes the robot t.o navigate around obstacles on the wa.y Lo a final LaJ•get. Obstacle avoidance is performed in a reactive manner by representing the objects and target. in the robot's environment. as Gaussian funct.ions. However, t.he influence of the GaussiaJJS is modulated dynamically on the basis of the robol's behavior ill a way that avoids problems with local minjma. The proposed module enables the robot to operate successfully with dHferent obstacle configurations, such as corridors, nla%cs, doors and even concave obstacles.