Coverage Path Planning (CPP) describes the process of generating robot trajectories that fully cover an area or volume. Applications are, amongst many others, mobile cleaning robots, lawn mowing robots or harvesting machines in agriculture. Many approaches and facets of this problem have been discussed in literature but despite the availability of several surveys on the topic there is little work on quantitative assessment and comparison of different coverage path planning algorithms. This paper analyzes six popular off-line coverage path planning methods, applicable to previously recorded maps, in the setting of indoor coverage path planning on room-sized units. The implemented algorithms are thoroughly compared on a large dataset of over 550 rooms with and without furniture.

The division of floor plans or navigation maps into single rooms or similarly meaningful semantic units is central to numerous tasks in robotics such as topological mapping, semantic mapping, place categorization, human-robot-interaction, or automatized professional cleaning. Although many map partitioning algorithms have been proposed for various applications there is a lack of comparative studies on these different algorithms. This paper surveys the literature on room segmentation and provides four publicly available implementations of popular methods, which target the semantic mapping domain and are tuned to yield segmentations into complete rooms. In an attempt to provide new users of such technologies guidance in the choice of map segmentation algorithm, those methods are compared qualitatively and quantitatively using several criteria. The evaluation is based on a novel compilation of 20 challenging floor plans.

Millions of office workplaces are cleaned by a surprisingly small group of cleaning workers every day, however, cleaning companies struggle to recruit enough personnel these days. One solution to this challenge is to schedule available professionals for demanding tasks while relieving them from simpler activities which are transferred to a robotic cleaning assistant. Two of such tasks are floor cleaning and waste disposal which account for 70% of the daily cleaning efforts. This paper presents the world's first autonomous cleaning robot prototype that masters both of these tasks and whose development was accompanied by the advice of a large cleaning company. Besides a detailed description of the overall system and its individual components an evaluation is provided based on real world experiments. The results indicate that both cleaning tasks can be solved at high quality but with potential for increased efficiency to meet the required performance. Hence, the paper concludes with a discussion on measures necessary for the development of a commercial prototype.

Accurate surface normal computation is one of the most basic and important tasks for 3d perception. While much progress has been made in speeding up normal estimation algorithms and improving their accuracy, a significant inaccuracy still remains even with modern implementations, which is the correct determination of surface normals close to non-differentiable surface edges. Current algorithms tend to amalgamate neighborhood points from independent surfaces yielding normals that neither fit well to the one nor the other surface. This paper introduces a fast and accurate 3d edge detection algorithm suitable to detect discontinuities both in depth and on surfaces with nearly 90% accuracy at rates beyond 30 Hz. Based on this method, we demonstrate how established normal estimation algorithms can be extended for edge-awareness. Additionally, a new edge-aware, fast, accurate, and robust normal estimation approach is described which exploits the data structures computed for 3d edge detection and estimates normals at 23 Hz. We assess the performance of all proposed methods and compare them with other state-of-the-art approaches.

Non-holonomic, omnidirectional undercarriages that are composed of steered standard wheels seem to provide a solid compromise between versatility, flexibility and high robustness against various ground conditions. However, such undercarriages are characterized by the occurrence of a number of singular configurations. To avoid these singular configurations most control-approaches restrict the admissible configuration-space thus eventually reducing the mobility and flexibility of the undercarriage. Within this work a state-space representation that forms a locally singularity-free atlas of the admissible configuration-space is presented. Based on this state-space description a switching based controller is developed that incorporates the former singular regions into the used configuration space and thus allows to exploit the full flexibility of non-holonomic, omnidirectional undercarriages. The implemented controller is quantitatively and qualitatively evaluated and compared to one approach that avoids the singular regions and one that completely neglects the non-holonomic bindings.

The integration of industrial robot systems into the manufacturing environments of small and medium sized enterprises is a key requirement to guarantee competitiveness and productivity. Due to the still complex and time-consuming procedure of robot path definition, novel programming strategies are needed, converting the robotic system into a flexible coworker that actively supports its operator. In this paper, a learning-from-demonstration strategy based on Hidden Markov Models is presented, which permits the robot system to adapt to user- as well as process-specific features. To evaluate the suitability of this approach for small-lot production, the learning strategy has been implemented for an arc welding robot and has been evaluated on-site at a medium sized metal-working company.

Currently, pseudo-omnidirectional, wheeled mobile robots with independently steered and driven wheels seem to provide a solid compromise between complexity, flexibility and robustness. Yet, such undercarriages are imposed to the risk of actuator fighting and suffer from singular regions within their configuration space. To address these problems we expand a previously developed potential field (PF) based approach by expanding it with a predictive horizon. The proposed method is based on a model predictive control (MPC) approach, incorporating a gradient descent optimization step via the Pontryagin minimum principle. To enforce adherence to the constraints during optimization, we modify the Lagrange-multipliers within the backpropagation of the costates. The proposed approach is evaluated simulatively w.r.t. the undercarriage of the Care-O-bot®3 mobile robot and is compared to the potential field based and a model predictive control approach.

For mobile platforms with steerable standard wheels it is necessary to precisely coordinate rotation and steering angle of their wheels. An established approach to ensure this is to represent the current state of motion in form of the Instantaneous Centre of Motion (ICM) and to derive a trajectory within this space. However, while control in the ICM space does guarantee adherence to the system's nonholonomic constraints, it does not avoid the system's singular configurations. Within this work we address the problem of singularity avoidance within the ICM space. Singularities related to the mathematical representation of the ICM are reduced by a reformulation of this representation. Furthermore, a controller based on artificial potential fields avoiding singular configurations of the robot by representing them as obstacles in the derived ICM space is designed. The resulting controller is particularized and analyzed w.r.t. the Care-O-bot 3 demonstrator.

This paper introduces Care-O-bot®3, a highly integrated and compact service robot with manipulation, navigation and vision capabilities. In particular, Care-O-bot®3 combines the best of available technology including a 7 DOF light-weight arm, an omnidirectional platform and many high-end sensors along with a sustainable, end user oriented design concept enabling many interaction possibilities.

Robots are emerging from industrial plants toward every people's daily life. Thus, navigation in and understanding of human related environments becomes a prerequisite for the systems of tomorrow. Most such environments can be efficiently described using line segments. However, incorporation of extent information is often difficult, as line segments are seldom observed completely and erroneous data-association may corrupt the information associated to a certain segment. To reduce such problems this paper proposes a statistically driven description of line segments. The corresponding parameters are decomposed into line and corner features, which are separately tracked through an extended Kalman filter (EKF). Information about the extent of the lines is encoded statistically. Therefore, we use a method to recursively incorporate the information gained through a time-series of measurements. Thus, the covariance matrix belonging to a line segment grows as new regions of the corresponding line are discovered. Experimental results obtained by implementation on the mobile platform ITrike show the validity of our algorithm.