Showing papers by "H. Levent Akin published in 2006"
TL;DR: The main aim is to find the best method which is very robust and fast and requires less computational power and memory compared to similar approaches and is accurate enough for high level decision making which is vital for robot soccer.
Abstract: In this work, several localization algorithms that are designed and implemented for Cerberus'05 Robot Soccer Team are analyzed and compared. These algorithms are used for global localization of autonomous mobile agents in the robotic soccer domain, to overcome the uncertainty in the sensors, environment and the motion model. The algorithms are Reverse Monte Carlo Localization (R-MCL), Simple Localization (S-Loc) and Sensor Resetting Localization (SRL). R-MCL is a hybrid method based on both Markov Localization (ML) and Monte Carlo Localization (MCL) where the ML module finds the region where the robot should be and MCL predicts the geometrical location with high precision by selecting samples in this region. S-Loc is another localization method where just one sample per percept is drawn, for global localization. Within this method another novel method My Environment (ME) is designed to hold the history and overcome the lack of information due to the drastically decrease in the number of samples in S-Loc. ME together with S-Loc is used in the Technical Challenges in Robocup 2005 and play an important role in ranking the First Place in the Challenges. In this work, these methods together with SRL, which is a widely used successful localization algorithm, are tested with both offline and real-time tests. First they are tested on a challenging data set that is used by many researches and compared in terms of error rate against different levels of noise, and sparsity. Besides time required recovering from kidnapping and speed of the methods are tested and compared. Then their performances are tested with real-time tests with scenarios like the ones in the Technical Challenges in ROBOCUP. The main aim is to find the best method which is very robust and fast and requires less computational power and memory compared to similar approaches and is accurate enough for high level decision making which is vital for robot soccer.
12 citations
01 Jan 2006
TL;DR: In this paper, a set of practical extensions to the vision-based Monte Carlo localization (MCL) for RoboCup Sony AIBO legged robot soccer domain is presented, in which two of them are novel approaches and the remaining ones are different from the previous implementations.
Abstract: This paper proposes a set of practical extensions to the vision-based Monte Carlo localization (MCL) for RoboCup Sony AIBO legged robot soccer domain. The main disadvantage of AIBO robots is that they have a narrow field of view so the number of landmarks seen in one frame is usually not enough for geometric calculation. MCL methods have been shown to be accurate and robust in legged robot soccer domain but there are some practical issues that should be handled in order to maintain stability/elasticity ratio in a reasonable level. In this work, we presented four practical extensions in which two of them are novel approaches and the remaining ones are different from the previous implementations.
11 citations
01 Jan 2006
TL;DR: In this article, a hybrid method based on both Markov Localization(ML) and Monte Carlo Localization (MCL) is proposed to overcome the uncertainty in the sensors, environment and the motion model.
Abstract: In this work, a novel method called Fuzzy Reverse Monte Carlo Localization (Fuzzy R-MCL) for global localization of autonomous mobile agents in the robotic soccer domain is proposed to overcome the uncertainty in the sensors, environment and the motion model. R-MCL is a hybrid method based on both Markov Localization(ML) and Monte Carlo Localization(MCL) where the ML module finds the region where the robot should be and MCL predicts the geometrical location with high precision by selecting samples in this region. In this work, a fuzzy approach is embedded in this method, to improve flexibility, accuracy and robustness. In addition to using Fuzzy membership functions in modeling the uncertainty of the grid cells and samples, different heuristics are used to enable the adaptation of the method to different levels of noise and sparsity. The method is very robust and fast and requires less computational power and memory compared to similar approaches and is accurate enough for high level decision making which is vital for robot soccer.
4 citations
TL;DR: Kalman Based Finite State Controller (KBFSC) constructs an internal world model over the continuous belief space, represented by a finite state automaton, representing the approximate optimal policy without the need for discretization of the state and observation space.
Abstract: A real world environment is often partially observable by the agents either because of noisy sensors or incomplete perception. Moreover, it has continuous state space in nature, and agents must decide on an action for each point in internal continuous belief space. Consequently, it is convenient to model this type of decision- making problems as Partially Observable Markov Decision Processes (POMDPs) with continuous observation and state space. Most of the POMDP methods whether approximate or exact assume that the underlying world dynamics or POMDP parameters such as transition and observation probabilities are known. However, for many real world environments it is very difficult if not impossible to obtain such information. We assume that only the internal dynamics of the agent, such as the actuator noise, interpretation of the sensor suite, are known. Using these internal dynamics, our algorithm, namely Kalman Based Finite State Controller (KBFSC), constructs an internal world model over the continuous belief space, represented by a finite state automaton. Constructed automaton nodes are points of the continuous belief space sharing a common best action and a common uncertainty level. KBFSC deals with continuous Gaussian-based POMDPs. It makes use of Kalman Filter for belief state estimation, which also is an efficient method to prune unvisited segments of the belief space and can foresee the reachable belief points approximately calculating the horizon N policy. KBFSC does not use an "explore and update" approach in the value calculation as TD-learning. Therefore KBFSC does not have an extensive exploration-exploitation phase. Using the MDP case reward and the internal dynamics of the agent, KBFSC can automatically construct the finite state automaton (FSA) representing the approximate optimal policy without the need for discretization of the state and observation space. Moreover, the policy always converges for POMDP problems.
3 citations
Journal Article•
TL;DR: A fuzzy approach is embedded in this method, to improve flexibility, accuracy and robustness, and different heuristics are used to enable the adaptation of the method to different levels of noise and sparsity.
Abstract: In this work, a novel method called Fuzzy Reverse Monte Carlo Localization (Fuzzy R-MCL) for global localization of autonomous mobile agents in the robotic soccer domain is proposed to overcome the uncertainty in the sensors, environment and the motion model. R-MCL is a hybrid method based on both Markov Localization(ML) and Monte Carlo Localization(MCL) where the ML module finds the region where the robot should be and MCL predicts the geometrical location with high precision by selecting samples in this region. In this work, a fuzzy approach is embedded in this method, to improve flexibility, accuracy and robustness. In addition to using Fuzzy membership functions in modeling the uncertainty of the grid cells and samples, different heuristics are used to enable the adaptation of the method to different levels of noise and sparsity. The method is very robust and fast and requires less computational power and memory compared to similar approaches and is accurate enough for high level decision making which is vital for robot soccer.