The following information was submitted:
Transactions: WSEAS TRANSACTIONS ON SYSTEMS AND CONTROL
Transactions ID Number: 32-253
Full Name: Ranka Kulic
Position: Assistant Professor
Age: ON
Sex: Female
Address: Bulevar Umetnosti 29, belgrade
Country: YUGOSLAVIA
Tel:
Tel prefix:
Fax:
E-mail address: rkulic2001@yahoo.com
Other E-mails: rkulic2008@gmail.com
Title of the Paper: Identification of Control Strategy to Avoid Unmoving and Moving Obstacles by Modification of Kohonen Rule
Authors as they appear in the Paper: Ranka Kuliæ, Zoran Vukiæ
Email addresses of all the authors: rkulic2001@yahoo.com, zoran.vukic@fer.hr
Number of paper pages: 16
Abstract: Abstract: - The problem of path generation for the autonomous vehicle in environments with the infinite number unmoving and moving obstacles is considered. Generally, the problem is known in the literature as the path planning. In this paper the algorithm, named MKBC, is connected with a characteristical idea of Kohonen rule and expiriance with RBF neuaral network. Among other things, Kohonen rule is connected with the weighting coefficients. The MKBC algorithm does not use the weighting values as values from the previous time, but permanentlly uses the training values as weighting values. This enables an intelligent system to learn from the examples (operator's demonstrations) to control a vehicle in avoiding obstacles, like the human operator does, which is known as behavioral cloning. Following the given context the problem narrow passage avoiding and the goal position reaching fundamentally is treated. Namely, defining if – then rule according to the!
problem the previously is observed as destroying of the consistency of the reached methodology. Then the problem the robot vehicle speed is treated in the manner to save consistency of the reached methodology. At the end, the robot distance holding according to the closest obstacle is also considered.The advantage of this approach lies in the fact that a complete path can be defined off-line, without using sophisticated symbolical models of obstacles. Important characteristic of the MKBC algorithm is polynomial complexity, while most other path planning algorithms are exponential. Experiments determined that it is robust to parameter change and suita¬ble for real time application. In the next phase it is expected to attempt to find optimal path.
Keywords: Behavioural clonning; Kohonen rule; neural network; robot vehicle path planning; nonlinear equation
EXTENSION of the file: .pdf
Special (Invited) Session: WSEAS Conference,Prague, CH
Organizer of the Session: Valeri M. Mladenov
How Did you learn about congress:
IP ADDRESS: 93.86.177.37
