Flavio Tonidandel

itSIMPLE: towards an integrated design system for real planning applications

Since the end of the 1990s there has been an increasing interest in the application of AI planning techniques to solve real-life problems. In addition to characteristics of academic problems, such as the need to reason about actions, real-life problems require detailed knowledge elicitation, engineering, and management. A systematic design process in which Knowledge and Requirements Engineering tools play a fundamental role is necessary in such applications. One of the main challenges in such design process, and consequently in the study of Knowledge Engineering in AI planning, has been the analysis of requirements and their subsequent transformation into an input-ready model for planners. itSIMPLE is a research project dedicated to the study of a project process to support the design phases of reallife planning models. In this paper, we give an overview of itSIMPLE focusing on the main translation processes among a minimal set of representations: from requirements represented in UML to Petri Nets and from UML models to PDDL for problem solving.

Case adaptation by segment replanning for case-based planning systems

An adaptation phase is crucial for a good and reasonable Case-Based Planning (CBP) system. The adaptation phase is responsible for finding a solution in order to solve a new problem. If the phase is not well designed, the CBP system may not solve the desirable range of problems or the solutions will not have appropriate quality. In this paper, a method called CASER – Case Adaptation by Segment Replanning – is presented as an adaptation rule for case-based planning system. The method has two phases: the first one completes a retrieved case as an easy-to-generate solution method. The second phase improves the quality of the solution by using a generic heuristic in a recursive algorithm to determine segments of the plan to be replanned. The CASER method does not use any additional knowledge, and it can find as good solutions as those found by the best generative planners.

Market-based dynamic task allocation using heuristically accelerated reinforcement learning

This paper presents a Multi-Robot Task Allocation (MRTA) system, implemented on a RoboCup Small Size League team, where robots participate of auctions for the available roles, such as attacker or defender, and use Heuristically Accelerated Reinforcement Learning to evaluate their aptitude to perform these roles, given the situation of the team, in real-time. The performance of the task allocation mechanism is evaluated and compared in different implementation variants, and results show that the proposed MRTA system significantly increases the team performance, when compared to pre-programmed team behavior algorithms.

On the construction of a RoboCup small size league team

The Robot Soccer domain has become an important artificial intelligence test bench and a widely studied research area. It is a domain with real, dynamic, and uncertain environment, where teams of robots cooperate and face adversarial competition. To build a RoboCup Small Size League (SSL) team able to compete in the world championship requires multidisciplinary research in fields like robotic hardware development, machine learning, multi-robot systems, computer vision, control theory, and mechanics, among others.This paper intends to provide insights about the aspects involved on the development of the RoboFEI RoboCup SSL robot soccer team and to present the contributions produced over its course. Among these contributions, a computer vision system employing an artificial neural network (ANN) to recognize colors, a heuristic algorithm to recognize partially detected objects, an implementation of the known rapidly-exploring random trees (RRT) path planning algorithm with additional rules, enabling the angle of approach of the robot to be controlled, and a layered strategy software system.Experimental results on real robots demonstrate the high performance of the vision system and the efficiency of the RRT algorithm implementation. Some strategy functions are also experimented, with empirical results showing their effectiveness.

Using Reinforcement Learning to Improve the Stability of a Humanoid Robot: Walking on Sloped Terrain

In order to perform a walk on a real environment, humanoid robots need to adapt themselves to the environment, as humans do. One approach to achieve this goal is to use Machine Learning techniques that allow robots to improve their behavior with time. In this paper, we propose a system that uses Reinforcement Learning to learn the action policy that will make a robot walk in an upright position, in a lightly sloped terrain. To validate this proposal, experiments were made with a humanoid robot - a robot for the RoboCup Humanoid League based on DARwIn-OP. The results showed that the robot was able to walk on sloping floors, going up and down ramps, even in situations where the slope angle changes.

Newton: A High Level Control Humanoid Robot for the RoboCup Soccer KidSize League

One of the goals of humanoid robot researchers is to develop a complete – in terms of hardware and software – artificial autonomous agent able to interact with humans and to act in the contemporary world, that is built for human beings. There has been an increasing number of humanoid robots in the last years, including Aldebaran’s NAO and Romeo, Intel’s Jimmy and Robotis’ DARwIn-OP. This research article describes the project and development of a new humanoid robot named Newton, made for research purposes and also to be used in the RoboCup Soccer KidSize League Competition. Newton robot’s contributions include that it has been developed to work without a dedicated microcontroller board, using an four-by-four-inch Intel NUC board, that is a fully functioning PC. To work with this high level hardware, a new software architecture comprised of completely independent processes was proposed. This architecture, called Cross Architecture, is comprised of completely independent processes, one for each intelligent system required by a soccer player: Vision, Localization, Decision, Communication, Planning, Sense and Acting, besides having a process used for managing the others. The experiments showed that the robot could walk, find the ball in an unknown position, recover itself from a fall and kicking the ball autonomously with a good performance.

Towards Model-based Vision Systems for Robot Soccer Teams

Since it’s beginning, Robot Soccer has been a platform for research and development of independent mobile robots and multi-agent systems, involving the most diverse areas of engineering and computer science. There are some problems to be solved in this domain, such as mechanical construction, electronics and control of mobile robots. But the main challenge is found in the areas related to Artificial Intelligence, as multi-agent systems, machine learning and computer vision. The problems and challenges mentioned above are not trivial, since Robot Soccer is dynamic, uncertain and probabilistic. A computer vision system for a Robot Soccer team must be fast and robust, and it is desirable that it can handle noise and luminous intensity variations. A number of techniques can be applied for object recognition in the domain of Robot Soccer, as described by (Grittani et al., 2000). The research of (Grittani et al., 2000) is based only on color information, as well as the research of (Weiss & Hildebrand, 2004) that uses color information to reduce the amount of information contained in each image frame through a called “relevance point filter”. Other researches uses the shape model of the objects to detect on the image, technique generally used in local vision systems. The research of (Gonner et al., 2005), for instance, detects the ball through it’s shape model projected on the image, a circumference, but still uses color-only information to recognize the robots. No matter which technique is used to solve the Robot Soccer computer vision challenge, it must be able to determine position and angle of the robots and the ball with maximum accuracy and minimal processing time possible, because the success of the strategy and control system depends on the information given by the computer vision system. This chapter extends the work presented by (Martins et al., 2006a), which considers the use of a well known image segmentation technique – the Hough Transform – to locate the mobile robots and the ball on global vision images, taking advantage of the domain characteristics – the robots and ball shape. To implement the Hough Transform technique, which is in most cases implemented in robotic systems using special hardware, only an offthe-shelf frame grabber and a personal computer are used. A new approach to interpret the Hough space is proposed, as well as the method used to recognize objects, which is based on a constraint satisfaction approach.

Reading PDDL, writing an object-oriented model

There are many efforts towards a combination of planning systems and real world applications. Although the PDDL is in constant evolution, which improves its capability to describe real domains, it is still a declarative language that is not so simple to be used by the non-planning community. This paper describes a translation process that reads a domain specification in PDDL and transforms it into an object-oriented model, more specifically into a version of UML for planning approaches. This translation process can let a designer read PDDL domains and verify it with some powerful tool like itSIMPLE or GIPO, or it can allow a planning system that only reads object-oriented models to run in domains described in PDDL originally.

Humanoid Robot Gait on Sloping Floors Using Reinforcement Learning

Climbing ramps is an important ability for humanoid robots: ramps exist everywhere in the world, such as in accessibility ramps and building entrances. This works proposes the use of Reinforcement Learning to learn the action policy that will make a robot walk in an upright position, in a lightly sloped terrain. The proposed architecture of our system is a two-layer combination of the traditional gait generation control loop with a reinforcement learning component. This allows the use of an accelerometer to generate a correction for the gait, when the slope of the floor where the robot is walking changes. Experiments performed on a real robot showed that the proposed architecture is a good solution for the stability problem.

A fast model-based vision system for a robot soccer team

Robot Soccer is a challenging research domain for Artificial Intelligence, which was proposed in order to provide a long-term problem in which researchers can investigate the construction of systems involving multiple agents working together in a dynamic, uncertain and probabilistic environment, to achieve a specific goal. This work focuses on the design and implementation of a fast and robust computer vision system for a team of small size robot soccer players. The proposed system combines artificial intelligence and computer vision techniques to locate the mobile robots and the ball, based on global vision images. To increase system performance, this work proposes a new approach to interpret the space created by a well-known computer vision technique called Hough Transform, as well as a fast object recognition method based on constraint satisfaction techniques. The system was implemented entirely in software using an off-the-shelf frame grabber. Experiments using real time image capture allows to conclude that the implemented system are efficient and robust to noises and lighting variation, being capable of locating all objects in each frame, computing their position and orientation in less than 20 milliseconds.