• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.15 no.4
• /
• pp.305-314
• /
• 2015

The work presented in this paper deals with a navigation problem for multiple mobile robot system in unknown indoor environments. The environment is completely unknown for all the robots and the surrounding information should be detected by the proximity sensors installed on the robots' bodies. In order to guide all the robots to move along collision-free paths and reach the goal positions, a navigation method based on the combination of a set of primary strategies has been developed. The indoor environments usually contain convex and concave obstacles. In this work, a danger judgment strategy in accordance with the sensors' data is used for avoiding small convex obstacles or moving objects which include both dynamic obstacles and other robots. For big convex obstacles or concave ones, a wall following strategy is designed for dealing with these special situations. In this paper, a state memorizing strategy is also proposed for the "infinite repetition" or "dead cycle" situations. Finally, when there is no collision risk, the robots will be guided towards the targets according to a target positioning strategy. Most of these strategies are achieved by the means of fuzzy logic controllers and uniformly applied for every robot. The simulation experiments verified that the proposed method has a positive effectiveness for the navigation problem.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2000.05a
• /
• pp.245-248
• /
• 2000

In this paper we improve the performance of autonomous mobile robot by induction of reinforcement learning concept. Generally, the system used in this paper is divided into two part. Namely, one is neural-fuzzy and the other is dynamic recurrent neural networks. Neural-fuzzy determines the next action of robot. Also, the neural-fuzzy is determined to optimal action internal reinforcement from dynamic recurrent neural network. Dynamic recurrent neural network evaluated to determine action of neural-fuzzy by external reinforcement signal from environment, Besides, dynamic recurrent neural network weight determined to internal reinforcement signal value is evolved by genetic algorithms. The architecture of propose system is applied to the computer simulations on controlling autonomous mobile robot.

• The Transactions of the Korean Institute of Electrical Engineers D
• /
• v.52 no.2
• /
• pp.85-92
• /
• 2003

In this paper, we propose a method to evaluate performances of mobile personal robots. A set of performance measures is proposed and the corresponding evaluation methods are developed. Different from industrial manipulators, personal robots need to be evaluated with its mobility, navigation, task and intelligent performance in environments where human beings exist. The proposed performance measures are composed of measures for mobility including vibration, repeatability, path accuracy and so on, as well as measures for navigation performance including wall following, overcoming doorsill, obstacle avoidance and localization. But task and intelligent behavior performances such as cleaning capability and high-level decision-making are not considered in this paper. To measure the proposed performances through a series of tests, we designed a test environment and developed measurement systems including a 3D Laser tracking system, a vision monitoring system and a vibration measurement system. We measured the proposed performances with a mobile robot to show the result as an example. The developed systems, which are installed at Korea Agency for Technology and Standards, are going to be used for many robot companies in Korea.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.15 no.5
• /
• pp.626-631
• /
• 2005

This paper presents the target object search algorithm under dynamic programming (DP) in the Tree-type maze. We organized an experimental environment with the concatenation Y-shape diverged way, small mobile robot, and a target object. By the principle of optimality, the backbone of DP, an agent recognizes that a given whole problem can be solved if the values of the best solution of certain ancillary problem can be determined according to the principle of optimality. In experiment, we used two different control algorithms: a left-handed method and DP. Finally we verified the efficiency of DP in the practical application using a real robot.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.21 no.3
• /
• pp.329-334
• /
• 2011

This paper presents the method for location estimation with obstacle tracking method. A laser scanner is used to implement the system, and we assume that the map information is known. We matches the measurement of the laser scanner to estimate the location of the robot by using sequential monte carlo (SMC) method. After estimating the robot's location, the pose of obstacles are detected and tracked, hence, we can predict the collision risk of them. Finally, we present the experiment results to verify the proposed method.

• Journal of the Institute of Convergence Signal Processing
• /
• v.3 no.2
• /
• pp.64-70
• /
• 2002

This paper proposes a new construction method of neural networks. The construction method consists of two fundmental ideas, which are a parallel selection-style evaluation and rules evolution. A new collision avoidance algorithm using genetic and neural network is proposed to avoid moving obstacles such as mobile robots. The input parameters of this algorithm is position of moving obstacles and target. Output is a regenerated direction of mobile robot. This algorithm is very simple and so, it is available to application of real time process. The pattern of collision avoidance is learned through test execution.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.37 no.2
• /
• pp.16-28
• /
• 2000

In this paper, we propose a navigation algorithm for a mobile robot, which is intelligently searching the goal location in unknown dynamic environments using an ultrasonic sensor. Instead of using "sensor fusion"method which generates the trajectory of a robot based upon the environment model and sensory data, "command fusion"method is used to govern the robot motions. The major factors for robot navigation are represented as a cost function. Using the data of the robot states and the environment, the weight value of each factor is determined for an optimal trajectory in dynamic environments. For the evaluation of the proposed algorithm, we peformed simulations in PC as well as real experiments with ZIRO. The results show that the proposed algorithm is apt to identify obstacles in unknown environments to guide the robot to the goal location safely.

• 제어로봇시스템학회:학술대회논문집
• /
• 2001.10a
• /
• pp.120.1-120
• /
• 2001

This paper proposes a prototype of a safe and reliable wheelchair robot with Human Robot Interaction (HRI). Since the wheelchair users are usually the handicapped, the wheelchair robot must guarantee the safety and reliability for the motion while considering users intention, A single color CCD camera is mounted for input user´s command based on human-friendly gestures, and a ultra sonic sensor array is used for sensing external motion environment. We use face and hand directional gestures as the user´s command. By combining the user´s command with the sensed environment configuration, the planner of the wheelchair robot selects an optimal motion. We implement a prototype wheelchair robot, MR, HURI (Mobile Robot with Human Robot Interaction) ...

• Journal of the Korean Institute of Intelligent Systems
• /
• v.13 no.1
• /
• pp.51-56
• /
• 2003

The main goal of existing mobile robot system was a complete autonomous navigation and the vision information was just used as an assistant way such as monitoring For this reason, the researches have been going towards sophistication of autonomousness gradually and the production costs also has been risen. However, it is also important to control remotely an inexpensive mobile robot system which has no intelligence at all. Such systems may be much more effective than fully autonomous systems in practice. Visual information from a simple camera and distance information from ultrasonic sensors are used for this system. Collision avoidance becomes the most important problem for this system. In this paper, we developed a force feedback joystick to control the robot system remotely with collision avoiding capability. Fuzzy logic is used for the algorithm in order to implement the expert s knowledge intelligently. Some experimental results show the force feedback joystick werks very well.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.24 no.1
• /
• pp.52-57
• /
• 2014

This paper presents an implementation of a smooth path planning method considering physical limits on a real time operating system for a two-wheel mobile robot. A Bezier curve is utilized to make a smooth path considering a robot's position and direction angle through the defined path. A convolution operator is used to generate the center velocity trajectory to travel the distance of the planned path while satisfying the physical limits. The joint space velocity is computed to drive the two-wheel mobile robot from the center velocity. Trajectory planning, velocity command according to the planned trajectory, and monitoring of encoder data are implemented with a multi-tasking system. And the synchronization of tasks is performed with a real-time mechanism of Event Flag. A real time system with multi-tasks is implemented and the result is compared with a non-real-time system in terms of path tracking to the designed path. The result shows the usefulness of a real-time multi-tasking system to the control system which requires real-time features.