As robots are gradually leaving highly structured factory environments and moving into human populated environments, they need to possess more complex cognitive abilities. They do not only have operate efficiently safely in natural, but also be able achieve higher levels of cooperation communication with humans. Human–robot collaboration (HRC) is a research field wide range applications, future scenarios, potentially high economic impact. HRC an interdisciplinary area comprising classical...

As mobile robotics is gradually moving towards a level of semantic environment understanding, robust 3D object recognition plays an increasingly important role. One the most crucial prerequisites for set fast algorithms geometry segmentation and extraction, which in turn rely on surface normal vectors as fundamental feature. Although there exists plethora different approaches estimating from point clouds, it largely unclear methods are preferable online processing robot. This paper presents...

This paper presents a novel online-capable interaction-aware intention and maneuver prediction framework for dynamic environments. The main contribution is the combination of model-based estimation with maneuver-based motion based on supervised learning. advantages this are twofold. On one hand, expert knowledge in form heuristics integrated, which simplifies modeling interaction. other difficulties associated scalability data sparsity algorithm due to so-called curse dimensionality can be...

This paper implements a real-time hand gesture recognition algorithm based on the inexpensive Kinect sensor. The use of depth sensor allows for complex 3D gestures where system is robust to disturbing objects or persons in background. A Haarlet-based implemented detect any orientation, and more particular pointing while extracting direction. integrated an interactive robot (based ROS), allowing interaction with robot. Pointing are translated into goals robot, telling him go. demo scenario...

In this paper we present a novel method for the efficient segmentation of 3D laser range data. The proposed algorithm is based on radially bounded nearest neighbor strategy and requires only two parameters. It yields deterministic, repeatable results does not depend any initialization procedure. efficiency verified with synthetic real

This paper presents a novel cooperative-driving prediction and planning framework for dynamic environments based on the methods of game theory. The proposed algorithm can be used highly automated driving highways or as sophisticated module advanced driver-assistance systems with no need intervehicle communication. main contribution this is model-based interaction-aware motion all vehicles in scene. In contrast to other state-of-the-art approaches, system also models replanning capabilities...

Automated vehicles require efficient and safe planning to maneuver in uncertain environments. Largely this uncertainty is caused by other traffic participants, e.g., surrounding vehicles. Future motion of often difficult predict. Whereas robust control approaches achieve safe, yet conservative for automated vehicles, Stochastic Model Predictive Control (SMPC) provides the presence uncertainty. Probabilistic constraints are applied ensure that maximal risk remains below a predefined level....

In this work, a framework for motion prediction of vehicles and safety assessment traffic scenes is presented. The developed can be used driver assistant systems as well autonomous driving applications. order to assess the future trajectories vehicle, these require all participants. As participants have mutual influence on each other, interaction them explicitly considered in framework, which inspired by an optimization problem. Taking into account, differs from existing approaches consider...

In robotics applications such as SLAM (Simultaneous Localization and Mapping), loop closure detection is an integral component required to build a consistent topological or metric map. This paper presents appearance based mechanism titled `IBuILD' (Incremental bag of BInary words for Appearance Loop Detection). The presented approach focuses on online, incremental formulation binary vocabulary generation detection. proposed does not require prior learning phase relies purely the scene...

Mapping and tracking in dynamic environments for autonomously-moving robots is still challenging, despite being essential tasks. They are often done separately using occupancy grids established object algorithms. In this work, an approach presented that estimates a uniform, low-level, grid-based world model including static objects, their uncertainties, as well velocities. It does not require existing tracks to filter out data points used creating updating the map. Nor it measurements can be...

Tracking and mapping the local environment form basis of an autonomous vehicle system. They are often realized separately using occupancy grids, which do not require object or shape assumptions, model-based tracking algorithms. Many approaches a binary classification sensor measurements into coming from static dynamic object, as otherwise inconsistencies between different representations likely to occur. This paper presents grid-based (GTAM), low-level approach that simultaneously estimates...

Modeling and estimating the current local environment by processing sensor measurement data is essential for intelligent vehicles. Static obstacles, dynamic objects, free space have to be appropriately represented, classified, filtered. Occupancy grids, known mapping static environments, provide a common low-level representation using occupancy probabilities with an implicit association through discrete grid structure. Extending this idea toward environments moving objects requires...

Several variations of methodological approaches are used to study the social acceptance in human-robot interaction. Due introduction robots home, working practice and usage typically informing design new forms technology missing. Studying interaction thus needs concepts. We propose a so called breaching experiment with additional ethnographic observation close this gap. To investigate concept we have been conducting field trial on public place. gathered feedback using questionnaires, order...

In this paper, the impact of facial expressions on HRI is explored. To determine their influence empathy a human towards robot and perceived subjective performance, an experimental setup created, in which participants engage dialog with head EDDIE. The web-based gaming application "Akinator" serves as backbone for structure. game, tries to guess thought-of person chosen by asking various questions about person. our evaluation, reacts ways human's expressions, either ignoring them, mirroring...

Being aware of mutual influences between individuals is a major requirement robot to efficiently operate in human populated environments. This especially true for the navigation among humans with its avoidance maneuvers. While easily manage this task, robotic systems are still facing problems. Most recent approaches concentrate on predicting motions individually and deciding afterwards. Thereby, interactivity mostly neglected. In work, we go one step back focus understanding underlying...

Robot motion planners are increasingly being equipped with an intriguing property: human likeness. This property can enhance human–robot interactions and is essential for a convincing computer animation of humans. paper presents (multi-agent) planner dynamic environments that generates human-like motion. The presented stands out against other by explicitly modeling decision making taking interdependencies between individuals into account, which achieved applying game theory. Non-cooperative...

This article is dedicated to developing an online collision detection and identification (CDI) scheme for human-collaborative robots. The composed of a signal classifier diagnosor, which monitors the sensory signals robot system, detects occurrence physical human–robot interaction, identifies its type within short period. In beginning, we conduct experiment construct data set that contains segmented interaction with ground truth. Then, develop on paradigm supervised learning. To adapt...

Motion planning algorithms for urban automated driving must handle uncertainty due to unknown intention and future motion of Dynamic Obstacles (DOs).Considering a single trajectory each DO is not adequate, especially in frameworks where traffic participants exhibit very different behaviors.However, including multiple candidate trajectories representing behaviors results an excess conservatism.We present novel combination the Interactive Multiple Model (IMM) algorithm Stochastic Predictive...

In mobile robotics, the segmentation of range data is an important prerequisite to object recognition and environment understanding. This paper presents algorithm for realtime a continuous stream incoming data. The method extension previously developed RBNN proceeds in two phases: Firstly, normal vector each point estimated from its neighborhood, which continuously monitored. Secondly, new points are clustered according their Euclidean angular distance points. An outline complexity as well...