Performance Comparison of Two Reinforcement Learning Algorithms for Small Mobile Robots
نویسندگان
چکیده
The design of intelligent agents by means of reinforcement learning is studied in this paper. A relational reinforcement learning algorithm is used to achieve a compact knowledge representation. Moreover, this approach allows to improve the learning performance by augmenting the algorithm with the so-called background knowledge. A case study on simulated physical robotic agents is performed and compared with our previous evolutionary robotics experiments in order to justify our approach.
منابع مشابه
Direct Optimal Motion Planning for Omni-directional Mobile Robots under Limitation on Velocity and Acceleration
This paper describes a low computational direct approach for optimal motion planning and obstacle avoidance of Omni-directional mobile robots within velocity and acceleration constraints on the robot motion. The main purpose of this problem is the minimization of a quadratic cost function while limitation on velocity and acceleration of robot is considered and collision with any obstacle in the...
متن کاملDelay Compensation on Fuzzy Trajectory Tracking Control of Omni-Directional Mobile Robots
This paper presents a delay compensator fuzzy control for trajectory tracking of omni-directional mobile robots. Fuzzy logic control (FLC) of the robots is a suitable strategy for dealing with model uncertainties, nonlinearities and disturbances. On the other hand, in many robotic applications such as mobile robots, delay phenomenon is able to substantially deteriorate the behavior of system's...
متن کاملLearning Algorithms for Small Mobile Robots: Case Study on Maze Exploration
An emergence of intelligent behavior within a simple robotic agent is studied in this paper. Two control mechanisms for an agent are considered — new direction of reinforcement learning called relational reinforcement learning, and a radial basis function neural network trained by evolutionary algorithm. Relational reinforcement learning is a new interdisciplinary approach combining logical pro...
متن کاملA Multiagent Reinforcement Learning algorithm to solve the Community Detection Problem
Community detection is a challenging optimization problem that consists of searching for communities that belong to a network under the assumption that the nodes of the same community share properties that enable the detection of new characteristics or functional relationships in the network. Although there are many algorithms developed for community detection, most of them are unsuitable when ...
متن کاملContinual Learning for Mobile Robots
Autonomous mobile robots should be able to learn incrementally and adapt to changes in the operating environment during their entire lifetime. This is referred to as continual learning. In this thesis, I propose an approach to continual learning which is based on adaptive state-space quantisation and reinforcement learning. Representational tools for continual learning should be constructive, a...
متن کامل