Frank Wallner

Real-time map refinement by fusing sonar and active stereo-vision

A good map of its environment is essential for efficient task execution of a mobile robot. Real time map update, especially in dynamic scenes is a difficult problem due to noisy sensor data and limited observation time. The paper describes a mapping procedure which identifies new obstacles in a scene and constructs a 3D surface model of it. This description is included in the geometrical map which robot navigation relies on. The mapping procedure is based on sonar range information and scenes reconstructed from stereo vision. The combination of sonar and stereo vision is advantageous, due to a complementary error characteristic concerning range and angular resolution. For sonar data integration the idea of local probability grids is proposed. Local grids which only cover areas where new obstacles are expected, reduce the complexity of grid based sonar data integration and can be applied to a dynamic environment. The partial models of an object that has been observed from different viewpoints are fused to a homogeneous description in a later step. A complex example shows the mapping procedure work robustly in dynamic indoor environments.

Position estimation using principal components of range data

Abstract This paper describes a new approach to mobile robot position estimation, based on principal component analysis of laser range data. An eigenspace is constructed from the principal components of a large number of range data sets. The structure of an environment, as seen by a range sensor, is represented as a family of surfaces in this space. Subsequent range data sets from the environment project as a point in this space. Associating this point to the family of surfaces gives a set of candidate positions and orientations (poses) for the sensor. These candidate poses correspond to positions and orientations in the environment which have similar range profiles. A Kalman filter can be used to select the most likely candidate pose based on coherence with small movements. The first part of this paper describes how a relatively small number of depth profiles of an environment can be used to generate a complete eigenspace. This space is used to build a representation of the range scan profiles obtained from a regular grid of positions and orientations (poses). This representation has the form of a family of surface (a manifold). This representation converts the problem of associating a range profile to possible positions and orientations into a table lookup. As a side benefit, the method provides a simple means to detect obstacles in a range profile. The final section of the paper reviews the use of estimation theory to determine the correct pose hypothesis by tracking.

Real-time map refinement by use of sonar and active stereo-vision

A mobile robot that navigates in an unknown or changing environment needs to maintain a dynamic model of its environment in order to update existing knowledge about the world. This paper addresses the problem of how to update a local model of the perceivable environment of a mobile robot and how to produce a description of newly perceived obstacles. A new hybrid concept is presented which combines the advantages of parameter- and grid-based modelling techniques. By introducing local grids that only cover areas where new obstacles are expected, the complexity of the map building process is significantly reduced. Local grids can be used in dynamic environment which was not possible with earlier grid-modelling approaches. The information, grids are constructed from, is provided by ultrasonic range sensors and an active stereo-vision system. Fusing information of both sensor systems is advantageous, especially due to their complementary error characteristics concerning range and angular resolution.

Integration of topological and geometrical planning in a learning mobile robot

The problem of adapting mobile robot navigation to changes in the environment is usually approached by modifying an internal world model. Descriptions on different levels of abstraction provide the information necessary for navigation and therefore influence the robots behaviour. The effect of such indirect adaptation is limited. The approach presented in this paper describes a new technique for direct integration of navigation experience in path planning. Thus, not only the world knowledge, but also the planning behaviour is improved over time. Experiments are carried out on a robot which is controlled by a layered architecture. It is integrated in a multirobot control environment which is described. The focus of the article is towards improving the higher navigation levels. The main idea being presented is the realization of adaptive behaviour not only on the level of reflexes, but also with respect to the planning capabilities of the robot. The application of learning techniques allows to continuously improve the estimation of plan costs and therefore the inherent strategy of the topological planner. It is illustrated that a combined learning of world description and navigation allows fast and sophisticated reaction to new environmental conditions.<<ETX>>

PRIAMOS: An experimental platform for reflexive navigation

Abstract In this paper the new mobile robot system PRIAMOS is presented. The system emphasizes on fast navigation algorithms. In a (partially) unknown or changing environment fast perception and reaction is necessary for efficient task execution. Low level mapping and planning modules are used to span the period until higher control layers have enough time to react. The first part of this paper briefly describes the mechanical and technical construction of PRIAMOS. After an explanation of the software architecture the solutions for motion control and reflexive navigation, as realized in PRIAMOS, are given.

Learning coordination skills in multi-agent systems

While distributed control architectures have many advantages over centralized ones, such as their inherent modularity and fault tolerance, a major problem of such architectures is to ensure the goal-oriented behaviour of the controlled system. This paper presents a framework within which the coordination skills required for goal-orientedness are learned from user demonstrations. The framework is based on a state-space model of the single agents building the system and a corresponding model of the coordination mechanism. Our mobile robot PRIAMOS provides an application example.

The Control Architecture of the Mobile System Priamos

Abstract In this paper the control architecture of the mobile robot system PRIAMOS is presented. It consists of four layers which integrate reactive and reflexive components in such a way that the robot can combine fast reaction to unexpected obstacles with predictive behaviour to known components in the environnement. A solution for the problem bow to combine control layers which process information on different levels of abstraction and of increasing complexity is discussed. The reflexive part of the control architecture and the modification concerning the motion controller of the mobile robot which was necessary to apply the idea of reflexive navigation is described in more details. The architecture presented allows to execute motion tasks in partially unknown or changing environment in an efficient manner. Especially the time critical reflexive level is computationally less expensive than previous approaches

Interactive control and environment modelling for a mobile robot based on multisensor perceptions

This paper presents an advanced telerobotic control system for a mobile robot with multisensor feedback. A telecontrol concept for various degrees of cooperation between a human operator and a mobile robot is described. With multisensor on-board the robot at the remote site can adjust its path while continuously accepting commands from the human operator. Interactive modelling that allows the modelling of an unknown environment and makes landmarks known to the robot is introduced. A graphical user interface and a 3-D animation system are important elements in the teleoperation, they are integrated in this system to help the operator by task analysis, off-line teaching and on-line monitoring. Experiments performed with the mobile robot PRIAMOS are discussed.

The control architecture of the priamos mobile system

Abstract In this paper the control architecture of the PRIAMOS mobile robot system is presented. It consists of four layers which integrate reactive and reflexive components in such a way that the robot can combine fast reaction to unexpected obstacles with predictive behaviour to known components in the environment. A solution for the problem of how to conbine control layers which process information on different levels of abstraction and of increasing complexity is discussed. The reflexive part of the control architecture and the modification concerning the motion controller of the mobile robot which was necessary to apply the idea of reflexive navigation is described in more detail. The architecture presented allows motion tasks to be executed in partially unknown or changing environment in an efficient manner. Especially, the time-critical reflexive level is computationally less expensive than in previous approaches.

Calibration of the active stereo vision system KASTOR with standardized perspective matrices

In this paper the new active stereo vision head of the IPR is presented. It is designed to serve as a flexible sensing device mounted on a mobile robot platform. KASTOR has eight motorized optical and mechanical degrees of freedom. The article describes the design of KASTOR as well as the real-time vision system it is connected to. Then a new head calibration technique is presented. The perspective matrices are computed directly without the need to determine internal or external camera parameters. This is achieved by the observation of a reference object in a scene. As soon as the robot moves any of its degrees of freedom, the calibration has to be updated. However it is difficult and time expensive to define a reference object when the head is mounted on a mobile robot. A solution is to compute all possible perspective matrices in advance and to store them in the memory. During autonomous operation the correct matrices are selected according to the head-configuration. A standardizing technique is presented, which reduces the amount of matrices to be stored significantly. Finally the article discusses the quality of calibration when in use as active sensing device on a mobile robot.