|
Group robot control in non-deterministic environments using the potential field method |
|
Igor Olegovich Shapovalov, Victor Vladimirovich Soloviev, Denis Aleksandrovich Beloglazov, Valery Ivanovich Finaev, Alexey Evgenyevich Titov(Southern Federal University, Russian Federation) |
| In this paper we describe the method of group robot control in non-deterministic environments. The stages of the solution of two interrelated tasks of robot group motion organization using the method of potential fields are considered: task of trajectory planning of the group considered as a uniform structure in the global coordinate space and the task of planning of the individual trajectory of each mobile robot in the considered structure. As main results, we propose the concept of structure formation for the robot groups using the space lattice and the algorithm of group robot control. |
|
|
Autonomous Robot Driving Decision Strategy Following Road Signs And Traffic Rules : Simulation Validation |
|
Zheng Hao Chong(Nanyang Technological University, Singapore), Ling Yee(Nidec Corporation, Singapore), Albert Causo, I Ming Chen(Nanyang Technological University, Singapore) |
| Autonomous robot driving decision strategy following road signs and traffic rules is described using simulation with a Turtlebot. A fully autonomous robot driving strategy is presented, which follows human level logic in decision making. A vision-based autonomous vehicles navigation system for road vehicles includes three main parts: 1) road and traffic signs detection; 2) vehicle movement guidance system; and 3) decision making following human level logic. A simulation world is built according to real environment scenario proving the concept of study. |
|
|
An Application of Pareto Solution with Adapted ACO for Searching Optimal Route of a Mobile Robot in Rough Terrain Environment |
|
Anuntapat Anuntachai, Arit Thammano, Olarn Wongwirat(King Mongkut's Institute of Technology Ladkrabang, Thailand) |
| This paper presents a novel method to find an optimal route of a mobile robot in rough terrain environment by using a Pareto solution with adapted ACO. The adapted ACO is used to search the routes containing the least energy and the shortest distance. Then, the Pareto solution is deployed to find the optimal route in terms of energy and distance by adopting a distance vector for selection. The zero and average values are used as references. The experimental results show that the proposed searching method can prescribe the optimal value for choosing the route provided by the proposed method. |
|
|
An Artificial Immune System for Fault Tolerant Control of an Over-Actuated Rover |
|
Robert Kidd(Suny Maritime College, United States), Carl Crane(University of Florida, United States) |
| This paper addresses the use of a subset of genetic algorithms, an artificial immune system, to increase the
responsiveness and accuracy of fault tolerant control in an over-actuated vehicle. These algorithms allow both inexact a priori to be easily incorporated into a model to represent the vehicle even within dynamic fault states. |
|
|
Improvement of optical flow obtained by compensation method using multilayer resolution structure |
|
Ryota Hirakawa(Graduate School of Science and Technology, Kumamoto University, Japan), Teruo Yamaguchi, Hiroshi Harada(Kumamoto University, Japan) |
| Optical flow is referred to as velocity distribution of object image,
and spatio-temporal differentiation method has been used to obtain optical flow.
It is possible to obtain optical flow fast by the method, but the high velocity may induce large systematic errors.
The compensation method has been introduced to reduce the magnitude of velocity
which is applied to spatio-temporal differentiation.
To determine the compensation velocity, the multilayer structure has been used.
Experimental results show that more accurate optical flow can be estimated in wider area. |
|
|
A Feasibility Study of Vehicle Pose Estimation using Road Sign Information |
|
Chanhee Jang, Young-Keun Kim(Handong University, Korea) |
| Feasibility test was studied on pose estimation algorithm using road sign information which is freely available online. The proposed method is to detect the upcoming road direction sign, compare it with DB image, and calculate the relative pose between the road sign and the vehicle. Since the absolute position of the sign post is provided by the database, the absolute position of the vehicle can be derived. This method is applicable to nationwide since the database stores all direction signs in the country. It can also be used to complement GPS signals in a navigation system. |
|