Dr. Armin Hornung - Publications

Publications

Show all abstracts Download all BibTex entries

You can also find my publications at Google Scholar and DBPL.

2014

Armin Hornung, Sebastian Böttcher, Christian Dornhege, Andreas Hertle, Jonas Schlagenhauf, and Maren Bennewitz
Mobile Manipulation in Cluttered Environments with Humanoids: Integrated Perception, Task Planning, and Action Execution
In: Proceedings of the IEEE-RAS International Conference on Humanoid Robots, 2014.
(Show abstract) (Hide abstract) (Show BibTeX) (Hide BibTeX) PDF Video

To autonomously carry out complex mobile manipulation tasks, a robot control system has to integrate several components for perception, world modeling, action planning and replanning, navigation, and manipulation. In this paper, we present a modular framework that is based on the Temporal Fast Downward Planner and supports external modules to control the robot. This allows to tightly integrate individual sub-systems with the high-level symbolic planner and enables a humanoid robot to solve challenging mobile manipulation tasks. In the work presented here, we address mobile manipulation with humanoids in cluttered environments, particularly the task of collecting objects and delivering them to designated places in a home-like environment while clearing obstacles out of the way. We implemented our system for a Nao humanoid tidying up a room, i.e., the robot has to collect items scattered on the floor, move obstacles out of its way, and deliver the objects to designated target locations. Despite the limited sensing and motion capabilities of the low-cost platform, the experiments show that our approach results in reliable task execution by applying monitoring actions to verify object and robot states.
```
@INPROCEEDINGS{hornung14humanoids,
  author = {Armin Hornung and Sebastian Boettcher and Christian Dornhege
  and Andreas Hertle and Jonas Schlagenhauf and Maren Bennewitz},
  title = {Mobile Manipulation in Cluttered Environments with Humanoids: 
  Integrated Perception, Task Planning, and Action Execution},
  booktitle = {Proc.~of the IEEE-RAS International Conference on Humanoid 
  Robots (HUMANOIDS)},
  year = 2014,
  month = {November},
  address = {Madrid, Spain},
}
```
Armin Hornung.
Humanoid Robot Navigation in Complex Indoor Environments.
PhD thesis, Albert-Ludwigs-Universität, Freiburg, Germany 2014.
(Show abstract) (Hide abstract) (Show BibTeX) (Hide BibTeX) PDF FreiDok

Humanoid service robots promise a high adaptability to environments designed for humans due to their human-like body layout. They are thus well-suited to assist humans in domestic environments for everyday tasks or providing elderly care, but also to replace human workers in hazardous environments. The human-like structure allows versatile manipulation with two arms as well as stepping over or onto obstacles with bipedal locomotion. Compared to wheeled platforms, however, the kinematic structure of humanoids requires active balancing while walking and allows only a limited payload, e.g., for additional sensors. Inaccurate motion execution can lead to foot slippage and thus to a poor estimation of ego-motion. Furthermore, the high number of degrees of freedom make planning and control a challenging problem for humanoid robots.

Autonomous navigation is a core capability of any service robot that performs high-level tasks involving mobility. Before a robot can fulfill a task, it needs to know where it is in the environment, what the environment looks like, and how it can successfully reach its goal. In this thesis, we present several novel contributions to the field of humanoid robotics with a focus on navigation in complex environments. We hereby cope with challenges in the areas of 3D environment representations, localization, perception, motion planning, and mobile manipulation.

As a basis, we first introduce a memory-efficient 3D environment representation along with techniques for building three-dimensional maps. In this representation, our probabilistic localization approach estimates the 6D pose of the humanoid based on data of its noisy onboard sensors. We compare different range sensors as well as sensor and motion models. For the critical task of climbing stairs, we develop an improved particle filter that additionally integrates vision data for a highly accurate pose estimate. We also present methods to perceive staircases in 3D range data. For reaching a navigation goal with a humanoid robot, we introduce and compare different search-based footstep planning approaches with anytime capabilities. We then investigate planning for the whole body of the humanoid while considering constraints such as maintaining balance during a manipulation task. This enables the robot to pick up objects or open doors. Finally, we present an efficient approach for navigation in three-dimensional cluttered environments that is particularly suited for mobile manipulation.

All techniques developed in this thesis were thoroughly evaluated both with real robots and in simulation. Our contributions generally advance the navigation capabilities of humanoid robots in complex indoor environments.
```
@PHDTHESIS{hornung14phd,
  author = {Armin Hornung},
  title = {Humanoid Robot Navigation in Complex Indoor Environments},
  school = {Albert-Ludwigs-Universit{\"a}t Freiburg},
  year = {2014},
  address = {Freiburg, Germany},
  month = {March}
}
```
Armin Hornung, Stefan Osswald, Daniel Maier, and Maren Bennewitz
Monte Carlo Localization for Humanoid Robot Navigation in Complex Indoor Environments
In: International Journal of Humanoid Robotics Volume 11, Issue 2, 2014.
(Show abstract) (Hide abstract) (Show BibTeX) (Hide BibTeX) PDF DOI: 10.1142/S0219843614410023

Accurate and reliable localization is a prerequisite for autonomously performing high-level tasks with humanoid robots. In this article, we present a probabilistic localization method for humanoid robots navigating in arbitrary complex indoor environments using only onboard sensing, which is a challenging task. Inaccurate motion execution of biped robots leads to an uncertain estimate of odometry, and their limited payload constrains perception to observations from lightweight and typically noisy sensors. Additionally, humanoids do not walk on flat ground only and perform a swaying motion while walking, which requires estimating a full 6D torso pose. We apply Monte Carlo localization to globally determine and track a humanoid's 6D pose in a given 3D world model, which may contain multiple levels and staircases. We present an observation model to integrate range measurements from a laser scanner or a depth camera as well as attitude data and information from the joint encoders. To increase the localization accuracy, e.g., while climbing stairs, we propose a further observation model and additionally use monocular vision data in an improved proposal distribution. We demonstrate the effectiveness of our methods in extensive real-world experiments with a Nao humanoid. As the experiments illustrate, the robot is able to globally localize itself and accurately track its 6D pose while walking and climbing stairs.
```
@ARTICLE{hornung14ijhr,
  author = {Armin Hornung, Stefan Osswald, Daniel Maier, Maren Bennewitz},
  title = {Monte Carlo Localization for Humanoid Robot Navigation in 
  Complex Indoor Environments},
  journal = {International Journal of Humanoid Robotics},
  volume = {11},
  number = {02},
  year = {2014},
  month= {June},
  doi = {10.1142/S0219843614410023},
}
```

2013

Armin Hornung
3D Mapping with OctoMap
Presentation at ROSCon 2013.
Slides (PDF)
Armin Hornung, Daniel Maier, and Maren Bennewitz
Search-Based Footstep Planning
In: Proceedings of the ICRA Workshop on Progress and Open Problems in Motion Planning and Navigation for Humanoids, 2013.
(Show abstract) (Hide abstract) (Show BibTeX) (Hide BibTeX) PDF

Efficient footstep planning for humanoid navigation through cluttered environments is still a challenging problem. Often, obstacles create local minima in the search space, forcing heuristic planners such as A* to expand large areas. Furthermore, planning longer footstep paths often takes a long time to compute. In this work, we introduce and discuss several solutions to these problems. For navigation, finding the optimal path initially is often not needed as it can be improved while walking. Thus, anytime search-based planning based on the anytime repairing A* or randomized A* search provides promising functionality. It allows to obtain efficient paths with provable suboptimality within short planning times. Opposed to completely randomized methods, anytime search-based planners generate paths that are goal-directed and guaranteed to be no more than a certain factor longer than the optimal solution. By adding new stepping capabilities and accounting for the whole body of the robot in the collision check, we extend the footstep planning approach to 3D. This enables a humanoid to step over clutter and climb onto obstacles. We thoroughly evaluated the performance of search-based planning in cluttered environments and for longer paths. We furthermore provide solutions to efficiently plan long trajectories using an adaptive level-of-detail planning approach.
```
@INPROCEEDINGS{hornung13icraws,
  author = {Armin Hornung and Daniel Maier and Maren Bennewitz},
  title = {Search-Based Footstep Planning},
  booktitle = {Proc.~of the ICRA Workshop on Progress and Open Problems in
  Motion Planning and Navigation for Humanoids},
  year = 2013,
  month = {May},
  address = {Karlsruhe, Germany}
}
```
Felix Burget, Armin Hornung, and Maren Bennewitz
Whole-Body Motion Planning for Manipulation of Articulated Objects
In: Proceedings of the IEEE International Conference on Robotics and Automation (ICRA), 2013.
(Show abstract) (Hide abstract) (Show BibTeX) (Hide BibTeX) PDF Video

Humanoid service robots performing complex object manipulation tasks need to plan whole-body motions that satisfy a variety of constraints: The robot must keep its balance, self-collisions and collisions with obstacles in the environment must be avoided and, if applicable, the trajectory of the end-effector must follow the constrained motion of a manipulated object in Cartesian space. These constraints and the high number of degrees of freedom make whole-body motion planning for humanoids a challenging problem. In this paper, we present an approach to whole-body motion planning with a focus on the manipulation of articulated objects such as doors and drawers. Our approach is based on rapidly-exploring random trees in combination with inverse kinematics and considers all required constraints during the search. Models of articulated objects hereby generate hand poses for sampled configurations along the trajectory of the object handle. We thoroughly evaluated our planning system and present experiments with a Nao humanoid opening a drawer, a door, and picking up an object. The experiments demonstrate the ability of our framework to generate solutions to complex planning problems and furthermore show that these plans can be reliably executed even on a low-cost humanoid platform.
```
@INPROCEEDINGS{burget13icra,
  author = {Felix Burget and Armin Hornung and Maren Bennewitz},
  title = {Whole-Body Motion Planning for Manipulation of Articulated Objects},
  booktitle = {Proc.~of the IEEE International Conference on Robotics and
  Automation (ICRA)},
  year = 2013,
  month = {May},
  address = {Karlsruhe, Germany}
}
```
Armin Hornung, Kai M. Wurm, Maren Bennewitz, Cyrill Stachniss, and Wolfram Burgard
OctoMap: An Efficient Probabilistic 3D Mapping Framework Based on Octrees
In: Autonomous Robots Vol. 34, Issue 3, 2013.
(Show abstract) (Hide abstract) (Show BibTeX) (Hide BibTeX) PDF DOI: 10.1007/s10514-012-9321-0 Project Website

Three-dimensional models provide a volumetric representation of space which is important for a variety of robotic applications including flying robots and robots that are equipped with manipulators. In this paper, we present an open-source framework to generate volumetric 3D environment models. Our mapping approach is based on octrees and uses probabilistic occupancy estimation. It explicitly represents not only occupied space, but also free and unknown areas. Furthermore, we propose an octree map compression method that keeps the 3D models compact. Our framework is available as an open-source C++ library and has already been successfully applied in several robotics projects. We present a series of experimental results carried out with real robots and on publicly available real-world datasets. The results demonstrate that our approach is able to update the representation efficiently and models the data consistently while keeping the memory requirement at a minimum.
```
@ARTICLE{hornung13auro,
  author = {Armin Hornung and Kai M. Wurm and Maren Bennewitz and Cyrill
  Stachniss and Wolfram Burgard},
  title = {{OctoMap}: An Efficient Probabilistic {3D} Mapping Framework Based
  on Octrees},
  journal = {Autonomous Robots},
  year = 2013,
  month= {April},
  url = {http://octomap.github.com},
  doi = {10.1007/s10514-012-9321-0},
  note = {Software available at \url{http://octomap.github.com}},
  volume = 34,
  issue = 3,
  pages = {189-206}
}
```

2012

Armin Hornung, Andrew Dornbush, Maxim Likhachev, and Maren Bennewitz.
Anytime Search-Based Footstep Planning with Suboptimality Bounds
In: Proceedings of the IEEE-RAS International Conference on Humanoid Robots (HUMANOIDS), 2012.
(Show abstract) (Hide abstract) (Show BibTeX) (Hide BibTeX) PDF

Efficient footstep planning for humanoid navigation through cluttered environments is still a challenging problem. Many obstacles create local minima in the search space, forcing heuristic planners such as A* to expand large areas. The goal of this work is to efficiently compute long, feasible footstep paths. For navigation, finding the optimal path initially is often not needed as it can be improved while walking. Thus, we propose anytime search-based planning using the anytime repairing A* (ARA*) and randomized A* (R*) planners. This allows to obtain efficient paths with provable suboptimality within short planning times. Opposed to completely randomized methods such as rapidly-exploring random trees (RRTs), these planners create paths that are goal-directed and guaranteed to be no more than a certain factor longer than the optimal solution. We thoroughly evaluated the planners in various scenarios using different heuristics. ARA* with the 2D Dijkstra heuristic yields fast and efficient solutions but its potential inadmissibility results in non-optimal paths for some scenarios. R*, on the other hand borrows ideas from RRTs, yields fast solutions, and is less dependent on a well-designed heuristic function. This allows it to avoid local minima and reduces the number of expanded states.
```
@INPROCEEDINGS{hornung12humanoids,
  author = {Armin Hornung and Andrew Dornbush and Maxim Likhachev and Maren 
  Bennewitz},
  title = {Anytime Search-Based Footstep Planning with Suboptimality Bounds},
  booktitle = {Proc.~of the IEEE-RAS International Conference on Humanoid 
  Robots (HUMANOIDS)},
  year = 2012,
  month = {November},
  address = {Osaka, Japan}
}
```
Daniel Maier, Armin Hornung, and Maren Bennewitz.
Real-Time Navigation in 3D Environments Based on Depth Camera Data
In: Proceedings of the IEEE-RAS International Conference on Humanoid Robots (HUMANOIDS), 2012.
(Show abstract) (Hide abstract) (Show BibTeX) (Hide BibTeX) PDF Video

In this paper, we present an integrated approach for robot localization, obstacle mapping, and path planning in 3D environments based on data of an onboard consumer-level depth camera. We rely on state-of-the-art techniques for environment modeling and localization, which we extend for depth camera data. We thoroughly evaluated our system with a Nao humanoid equipped with an Asus Xtion Pro Live depth camera on top of the humanoid's head and present navigation experiments in a multi-level environment containing static and non-static obstacles. Our approach performs in real-time, maintains a 3D environment representation, and estimates the robot's pose in 6D. As our results demonstrate, the depth camera is well-suited for robust localization and reliable obstacle avoidance in complex indoor environments.
```
@INPROCEEDINGS{maier12humanoids,
  author = {Daniel Maier and Armin Hornung and Maren Bennewitz},
  title = {Real-Time Navigation in {3D} Environments Based on Depth Camera Data},
  booktitle = {Proc.~of the IEEE-RAS International Conference on Humanoid 
  Robots (HUMANOIDS)},
  year = 2012,
  month = {November},
  address = {Osaka, Japan}
}
```
Christian Lutz, Felix Atmanspacher, Armin Hornung, and Maren Bennewitz.
NAO Walking Down a Ramp Autonomously
In: Video Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2012.
(Show abstract) (Hide abstract) (Show BibTeX) (Hide BibTeX) PDF Video

To fulfill high-level tasks, humanoid service robots must be able to autonomously and robustly navigate in man-made environments. These environments can be arbitrarily complex, containing multiple levels and various types of stairs or ramps connecting them. We previously presented techniques for autonomously climbing spiral staircases with humanoid robots. In this work, we extend this research direction by walking down ramps. On a Nao humanoid, we apply kinesthetic teaching to learn single stepping motions for the ramp. As we show in the experiments, by using the learned motions and integrating monocular vision and inertial data, the Nao is able to autonomously walk down a 2.10 m long ramp at an inclination of 20 degrees. The accompanying video shows the complete process of locating the beginning of the ramp using visual observations, walking down with regular corrections based on the inertial data, and finally determining the end of the ramp by detecting the ending edge before exiting the ramp.
```
@INPROCEEDINGS{lutz12iros,
  author =   {Christian Lutz and Felix Atmanspacher and Armin Hornung 
  and Maren Bennewitz},
  title =  {NAO Walking Down a Ramp Autonomously},
  booktitle =  {Video Proc.~of the IEEE/RSJ International Conference on
  Intelligent Robots and Systems (IROS)},
  year = 2012,
  month = {October},
  address = {Vilamoura, Portugal}
}
```
Stefan Oßwald, Armin Hornung, Maren Bennewitz.
Improved Proposals for Highly Accurate Localization Using Range and Vision Data
In: Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2012.
(Show abstract) (Hide abstract) (Show BibTeX) (Hide BibTeX) PDF Video

In order to successfully climb challenging staircases that consist of many steps and contain difficult parts, humanoid robots need to accurately determine their pose. In this paper, we present an approach that fuses the robot's observations from a 2D laser scanner, a monocular camera, an inertial measurement unit, and joint encoders in order to localize the robot within a given 3D model of the environment. We develop an extension to standard Monte Carlo localization (MCL) that draws particles from an improved proposal distribution to obtain highly accurate pose estimates. Furthermore, we introduce a new observation model based on chamfer matching between edges in camera images and the environment model. We thoroughly evaluate our localization approach and compare it to previous techniques in real-world experiments with a Nao humanoid. The results show that our approach significantly improves the localization accuracy and leads to a considerably more robust robot behavior. Our improved proposal in combination with chamfer matching can be generally applied to improve a range-based pose estimate by a consistent matching of lines obtained from vision.
```
@INPROCEEDINGS{osswald12iros,
  author =   {Stefan O{\ss}wald and Armin Hornung and Maren Bennewitz},
  title =  {Improved Proposals for Highly Accurate Localization Using Range 
  and Vision Data},
  booktitle =  {Proc.~of the IEEE/RSJ International Conference on
  Intelligent Robots and Systems (IROS)},
  year = 2012,
  month = {October},
  address = {Vilamoura, Portugal}
}
```
Armin Hornung
ROS for Humanoid Robots
Presentation at ROSCon 2012.
Slides (PDF)
Armin Hornung and Maren Bennewitz.
Adaptive Level-of-Detail Planning for Efficient Humanoid Navigation.
In: Proceedings of the IEEE International Conference on Robotics and Automation (ICRA), 2012.
(Show abstract) (Hide abstract) (Show BibTeX) (Hide BibTeX) PDF

In this paper, we consider the problem of efficient path planning for humanoid robots by combining grid-based 2D planning with footstep planning. In this way, we exploit the advantages of both frameworks, namely fast planning on grids and the ability to find solutions in situations where grid-based planning fails. Our method computes a global solution by adaptively switching between fast grid-based planning in open spaces and footstep planning in the vicinity of obstacles. To decide which planning framework to use, our approach classifies the environment into regions of different complexity with respect to the traversability. Experiments carried out in a simulated office environment and with a Nao humanoid show that (i) our approach significantly reduces the planning time compared to pure footstep planning and (ii) the resulting plans are almost as good as globally computed optimal footstep paths.
```
@INPROCEEDINGS{hornung12icra,
  author = {Armin Hornung and Maren Bennewitz},
  title = {Adaptive Level-of-Detail Planning for Efficient Humanoid Navigation},
  booktitle = {Proc.~of the IEEE International Conference on Robotics and 
  Automation (ICRA)},
  year = {2012},
  month = {May},
  address = {St. Paul, MN, USA}
}
```
Armin Hornung, Mike Phillips, E. Gil Jones, Maren Bennewitz, Maxim Likhachev, and Sachin Chitta.
Navigation in Three-Dimensional Cluttered Environments for Mobile Manipulation.
In: Proceedings of the IEEE International Conference on Robotics and Automation (ICRA), 2012.
Also presented at the RSS 2012 Workshop on Robots in Clutter: Manipulation, Perception and Navigation in Human Environments
(Show abstract) (Hide abstract) (Show BibTeX) (Hide BibTeX) PDF Video

Collision-free navigation in cluttered environments is essential for any mobile manipulation system. Traditional navigation systems have relied on a 2D grid map projected from a 3D representation for efficiency. This approach, however, prevents navigation close to objects in situations where projected 3D configurations are in collision within the 2D grid map even if actually no collision occurs in the 3D environment. Accordingly, when using such a 2D representation for planning paths of a mobile manipulation robot, the number of planning problems which can be solved is limited and suboptimal robot paths may result. We present a fast, integrated approach to solve path planning in 3D using a combination of an efficient octree-based representation of the 3D world and an anytime search-based motion planner. Our approach utilizes a combination of multi-layered 2D and 3D representations to improve planning speed, allowing the generation of almost real-time plans with bounded sub-optimality. We present extensive experimental results with the two-armed mobile manipulation robot PR2 carrying large objects in a highly cluttered environment. Using our approach, the robot is able to efficiently plan and execute trajectories while transporting objects, thereby often moving through demanding, narrow passageways.
```
@INPROCEEDINGS{hornung12icra_pr2,
  author = {Armin Hornung and Mike Phillips and E. Gil Jones and 
  Maren Bennewitz and Maxim Likhachev and Sachin Chitta},
  title = {Navigation in Three-Dimensional Cluttered Environments 
  for Mobile Manipulation},
  booktitle = {Proc.~of the IEEE International Conference on Robotics and 
  Automation (ICRA)},
  year = {2012},
  month = {May},
  address = {St. Paul, MN, USA}
}
```

2011

Maren Bennewitz, Daniel Maier, Armin Hornung, and Cyrill Stachniss.
Integrated Perception and Navigation in Complex Indoor Environments.
In: Proceedings of the HUMANOIDS 2011 workshop on Humanoid service robot navigation in crowded and dynamic environments, 2011.
(Show BibTeX)

@INPROCEEDINGS{bennewitz11humws,
  author =   {Maren Bennewitz and Daniel Maier and Armin Hornung and 
  Cyrill Stachniss},
  title = {Integrated Perception and Navigation in Complex Indoor Environments},
  booktitle = {Proc. of the HUMANOIDS 2011 workshop on Humanoid service 
  robot navigation in crowded and dynamic environments},
  year = 2011,
  month = {October},
  address = {Bled, Slovenia}
}

Stefan Oßwald, Jens-Steffen Gutmann, Armin Hornung, and Maren Bennewitz.
From 3D Point Clouds to Climbing Stairs: A Comparison of Plane Segmentation Approaches for Humanoids.
In: Proceedings of the IEEE-RAS International Conference on Humanoid Robots (HUMANOIDS), 2011.
(Show abstract) (Hide abstract) (Show BibTeX) (Hide BibTeX) PDF Video

In this paper, we consider the problem of building 3D models of complex staircases based on laser range data acquired with a humanoid. These models have to be sufficiently accurate to enable the robot to reliably climb up the staircase. We evaluate two state-of-the-art approaches to plane segmentation for humanoid navigation given 3D range data about the environment. The first approach initially extracts line segments from neighboring 2D~scan lines, which are successively combined if they lie on the same plane. The second approach estimates the main directions in the environment by randomly sampling points and applying a clustering technique afterwards to find planes orthogonal to the main directions. We propose extensions for this basic approach to increase the robustness in complex environments which may contain a large number of different planes and clutter. In practical experiments, we thoroughly evaluate all methods using data acquired with a laser-equipped Nao robot in a multi-level environment. As the experimental results show, the reconstructed 3D models can be used to autonomously climb up complex staircases.
```
@INPROCEEDINGS{osswald11humanoids,
  author = {Stefan O{\ss}wald and Jens-Steffen Gutmann and Armin Hornung 
  and Maren Bennewitz},
  title = {From 3{D} Point Clouds to Climbing Stairs: A Comparison of Plane 
  Segmentation Approaches for Humanoids},
  booktitle = {Proc.~of the IEEE-RAS International Conference 
  on Humanoid Robots (Humanoids)},
  year = 2011,
  month = {October},
  address = {Bled, Slovenia}
}    
```

Armin Hornung, E. Gil Jones, Sachin Chitta, Maren Bennewitz, Mike Phillips, and Maxim Likhachev.
Towards Navigation in Three-Dimensional Cluttered Environments.
In: Abstract Proceedings of The PR2 Workshop: Results, Challenges and Lessons Learned in Advancing Robots with a Common Platform at IROS 2011.
(Show BibTeX)

@INPROCEEDINGS{hornung11irosws,
  author = {A. Hornung and E. G. Jones and S. Chitta and  M. Bennewitz and 
  M. Phillips and M. Likhachev},
  title = {Towards Navigation in Three-Dimensional Cluttered Environments},
  booktitle = {Proc. of the IROS 2011 PR2 Workshop: Results, Challenges and 
  Lessons Learned in Advancing Robots with a Common Platform },
  year = 2011,
  month = {September},
  address = {San Francisco, CA, USA}  
}

Stefan Oßwald, Attila Görög, Armin Hornung, and Maren Bennewitz.
Autonomous Climbing of Spiral Staircases with Humanoids.
In: Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2011.
(Show abstract) (Hide abstract) (Show BibTeX) (Hide BibTeX) PDF Video

In this paper, we present an approach to enable a humanoid robot to autonomously climb up spiral staircases. This task is substantially more challenging than climbing straight stairs since careful repositioning is needed. Our system globally estimates the pose of the robot, which is subsequently refined by integrating visual observations. In this way, the robot can accurately determine its relative position with respect to the next step. We use a 3D~model of the environment to project edges corresponding to stair contours into monocular camera images. By detecting edges in the images and associating them to projected model edges, the robot is able to accurately locate itself towards the stairs and to climb them. We present experiments carried out with a Nao humanoid equipped with a 2D~laser range finder for global localization and a low-cost monocular camera for short-range sensing. As we show in the experiments, the robot reliably climbs up the steps of a spiral staircase.
```
@INPROCEEDINGS{osswald11iros,
  author = {Stefan O{\ss}wald and Attila G{\"o}r{\"o}g 
   and Armin Hornung and Maren Bennewitz},
  title = {Autonomous Climbing of Spiral Staircases with Humanoids},
  booktitle = {Proc.~of the IEEE/RSJ International Conference on
  Intelligent Robots and Systems (IROS)},
  year = 2011,
  month = {September},
  address = {San Francisco, CA, USA}
}
```
Johannes Garimort, Armin Hornung, and Maren Bennewitz.
Humanoid Navigation with Dynamic Footstep Plans.
In: Proceedings of the IEEE International Conference on Robotics and Automation (ICRA), 2011.
(Show abstract) (Hide abstract) (Show BibTeX) (Hide BibTeX) PDF Video Source code (ROS package)

Humanoid robots possess the capability of stepping over or onto objects, which distinguishes them from wheeled robots. When planning paths for humanoids, one therefore should consider an intelligent placement of footsteps instead of choosing detours around obstacles. In this paper, we present an approach to optimal footstep planning for humanoid robots. Since changes in the environment may appear and a humanoid may deviate from its originally planned path due to imprecise motion execution or slippage on the ground, the robot might be forced to dynamically revise its plans. Thus, efficient methods for planning and replanning are needed to quickly adapt the footstep paths to new situations. We formulate the problem of footstep planning so that it can be solved with the incremental heuristic search method D* Lite and present our extensions, including continuous footstep locations and efficient collision checking for footsteps. In experiments in simulation and with a real Nao humanoid, we demonstrate the effectiveness of the footstep plans computed and revised by our method. Additionally, we evaluate different footstep sets and heuristics to identify the ones leading to the best performance in terms of path quality and planning time. Our D* Lite algorithm for footstep planning is available as open source implementation.
```
@INPROCEEDINGS{garimort11icra,
  author = {Johannes Garimort and Armin Hornung and Maren Bennewitz},
  title = {Humanoid Navigation with Dynamic Footstep Plans},
  booktitle = {Proc.~of the IEEE International Conference on Robotics and 
  Automation (ICRA)},
  year = 2011,
  month = {May},
  address = {Shanghai, China}
}
```

2010

Armin Hornung, Kai M. Wurm, and Maren Bennewitz.
Humanoid Robot Localization in Complex Indoor Environments.
In: Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2010.
(Show abstract) (Hide abstract) (Show BibTeX) (Hide BibTeX) PDF Video

In this paper, we present a localization method for humanoid robots navigating in arbitrary complex indoor environments using only onboard sensing. Reliable and accurate localization for humanoid robots operating in such environments is a challenging task. First, humanoids typically execute motion commands rather inaccurately and odometry can be estimated only very roughly. Second, the observations of the small and lightweight sensors of most humanoids are seriously affected by noise. Third, since most humanoids walk with a swaying motion and can freely move in the environment, e.g., they are not forced to walk on flat ground only, a 6D torso pose has to be estimated. We apply Monte Carlo localization to globally determine and track a humanoid's 6D pose in a 3D world model, which may contain multiple levels connected by staircases. To achieve a robust localization while walking and climbing stairs, we integrate 2D laser range measurements as well as attitude data and information from the joint encoders. We present simulated as well as real-world experiments with our humanoid and thoroughly evaluate our approach. As the experiments illustrate, the robot is able to globally localize itself and accurately track its 6D pose over time.
```
@INPROCEEDINGS{hornung10iros,
  author = {Armin Hornung and Kai M. Wurm and Maren Bennewitz},
  title = {Humanoid Robot Localization in Complex Indoor Environments},
  booktitle =  {Proc.~of the IEEE/RSJ International Conference on Intelligent 
  Robots and Systems (IROS)},
  year = 2010,
  address = {Taipei, Taiwan},
  month= {October}
}
```
Armin Hornung, Maren Bennewitz, and Wolfram Burgard.
Learning Efficient Vision-based Navigation.
In: Abstract Proceedings of the International Conference on Indoor Positioning and Indoor Navigation (IPIN), 2010.
Armin Hornung, Maren Bennewitz, Cyrill Stachniss, Hauke Strasdat, Stefan Oßwald, and Wolfram Burgard.
Learning Adaptive Navigation Strategies for Resource-Constrained Systems.
In: Proceedings of the 3rd International Workshop on Evolutionary and Reinforcement Learning for Autonomous Robot Systems (ERLARS) at the European Conference on Artificial Intelligence, 2010.
(Show abstract) (Hide abstract) (Show BibTeX) (Hide BibTeX) PDF

The majority of navigation algorithms for mobile robots assume that the robots possess enough computational or memory resources to carry out the necessary calculations. Especially small and lightweight devices, however, are resource-constrained and have only restricted capabilities. In this paper, we present a reinforcement learning approach for mobile robots that considers the imposed constraints on their sensing capabilities and computational resources, so that they can reliably and efficiently fulfill their navigation tasks. Our technique learns a policy that optimally trades off the speed of the robot and the uncertainty in the observations imposed by its movements. It furthermore enables the robot to learn an efficient landmark selection strategy to compactly model the environment. We describe extensive simulated and real-world experiments carried out with both wheeled and humanoid robots which demonstrate that our learned navigation policies significantly outperform strategies using advanced and manually optimized heuristics.
```
@INPROCEEDINGS{hornung10erlars,
  author = {Armin Hornung and Maren Bennewitz and Cyrill Stachniss and 
  Hauke Strasdat and Stefan O{\ss}wald and Wolfram Burgard},
  title = {Learning Adaptive Navigation Strategies for Resource-constrained 
  Systems},
  booktitle = {Proc.~of the 3rd Int.~Workshop on Evolutionary and Reinforcement 
  Learning for Autonomous Robot Systems (ERLARS)},
  address = {Lisbon, Portugal},
  year = 2010,
  month = {August}
}
```
Armin Hornung and Maren Bennewitz.
Robust and Adaptive Navigation with Humanoid Robots.
In: Abstract Proceedings of the Workshop on Motion Planning: From Theory to Practice at Robotics: Science and Systems (RSS), 2010.
Armin Hornung, Maren Bennewitz, and Hauke Strasdat.
Efficient Vision-based Navigation - Learning about the Influence of Motion Blur.
In: Autonomous Robots, Vol. 29, Number 2, 2010.
(Show abstract) (Hide abstract) (Show BibTeX) (Hide BibTeX) PDF Video DOI: 10.1007/s10514-010-9190-3

In this article, we present a novel approach to learning efficient navigation policies for mobile robots that use visual features for localization. As fast movements of a mobile robot typically introduce inherent motion blur in the acquired images, the uncertainty of the robot about its pose increases in such situations. As a result, it cannot be ensured anymore that a navigation task can be executed efficiently since the robot's pose estimate might not correspond to its true location. We present a reinforcement learning approach to determine a navigation policy to reach the destination reliably and, at the same time, as fast as possible. Using our technique, the robot learns to trade off velocity against localization accuracy and implicitly takes the impact of motion blur on observations into account. We furthermore developed a method to compress the learned policy via a clustering approach. In this way, the size of the policy representation is significantly reduced, which is especially desirable in the context of memory-constrained systems. Extensive simulated and real-world experiments carried out with two different robots demonstrate that our learned policy significantly outperforms policies using a constant velocity and more advanced heuristics. We furthermore show that the policy is generally applicable to different indoor and outdoor scenarios with varying landmark densities as well as to navigation tasks of different complexity.
```
@ARTICLE{hornung10ar,
  author = {Armin Hornung and Maren Bennewitz and Hauke Strasdat},
  title = {Efficient Vision-based Navigation -- {L}earning about the Influence 
  of Motion Blur},
  journal = {Autonomous Robots},
  year = 2010,
  volume = 29,
  issue = 2,
  pages = {137-149},
  doi = {http://dx.doi.org/10.1007/s10514-010-9190-3}
}
```
Kai M. Wurm, Armin Hornung, Maren Bennewitz, Cyrill Stachniss, and Wolfram Burgard.
OctoMap: A Probabilistic, Flexible, and Compact 3D Map Representation for Robotic Systems.
In: Proceedings of the ICRA 2010 Workshop on Best Practice in 3D Perception and Modeling for Mobile Manipulation, 2010.
(Show abstract) (Hide abstract) (Show BibTeX) (Hide BibTeX) PDF Project Website

In this paper, we present an approach for modeling 3D environments based on octrees using a probabilistic occupancy estimation. Our technique is able to represent full 3D models including free and unknown areas. It is available as an open-source library to facilitate the development of 3D mapping systems. We also provide a detailed review of existing approaches to 3D modeling. Our approach was thoroughly evaluated using different real-world and simulated datasets. The results demonstrate that our approach is able to model the data probabilistically while, at the same time, keeping the memory requirement at a minimum.
```
@inproceedings{wurm10icraws,
author = {K. M. Wurm and A. Hornung and 
M. Bennewitz and C. Stachniss and W. Burgard},
title = {{OctoMap}: A Probabilistic, Flexible, and Compact {3D} Map 
Representation for Robotic Systems},
booktitle = {Proc. of the ICRA 2010 Workshop on Best Practice in 
3D Perception and Modeling for Mobile Manipulation},
year = 2010,
month = may,
address = {Anchorage, AK, USA},
url = {http://octomap.github.com},
note = {Software available at \url{http://octomap.github.com}}
}
```
Stefan Oßwald, Armin Hornung, and Maren Bennewitz.
Learning Reliable and Efficient Navigation with a Humanoid.
In: Proceedings of the IEEE International Conference on Robotics and Automation (ICRA), 2010.
(Show abstract) (Hide abstract) (Show BibTeX) (Hide BibTeX) PDF Video

Reliable and efficient navigation with a humanoid robot is a difficult task. First, the motion commands are executed rather inaccurately due to backlash in the joints or foot slippage. Second, the observations are typically highly affected by noise due to the shaking behavior of the robot. Thus, the localization performance is typically reduced while the robot moves and the uncertainty about its pose increases. As a result, the reliable and efficient execution of a navigation task cannot be ensured anymore since the robot's pose estimate might not correspond to the true location. In this paper, we present a reinforcement learning approach to select appropriate navigation actions for a humanoid robot equipped with a camera for localization. The robot learns to reach the destination reliably and as fast as possible, thereby choosing actions to account for motion drift and trading off velocity in terms of fast walking movements against accuracy in localization. We present extensive simulated and practical experiments with a humanoid robot and demonstrate that our learned policy significantly outperforms a hand-optimized navigation strategy.
```
@INPROCEEDINGS{osswald10icra,
  author = {Stefan O{\ss}wald and Armin Hornung and Maren Bennewitz},
  title = {Learning Reliable and Efficient Navigation with a Humanoid},
  booktitle = {Proc.~of the IEEE International Conference on Robotics and 
  Automation (ICRA)},
  year = {2010},
  address = {Anchorage, AK, USA},
  month = {May}
}
```

2009

Armin Hornung, Hauke Strasdat, Maren Bennewitz, and Wolfram Burgard.
Learning Efficient Policies for Vision-based Navigation.
In: Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2009.
(Show abstract) (Hide abstract) (Show BibTeX) (Hide BibTeX) PDF

Cameras are popular sensors for robot navigation tasks such as localization as they are inexpensive, lightweight, and provide rich data. However, fast movements of a mobile robot typically reduce the performance of vision-based localization systems due to motion blur. In this paper, we present a reinforcement learning approach to choose appropriate velocity profiles for vision-based navigation. The learned policy minimizes the time to reach the destination and implicitly takes the impact of motion blur on observations into account. To reduce the size of the resulting policies, which is desirable in the context of memory-constrained systems, we compress the learned policy via a clustering approach. Extensive simulated and real-world experiments demonstrate that our learned policy significantly outperforms any policy that uses a constant velocity. We furthermore show, that our policy is applicable to different environments. Additional experiments demonstrate that our compressed policies do not result in a performance loss compared to the originally learned policy.
```
@INPROCEEDINGS{hornung09iros,
  author = {Armin Hornung and Hauke Strasdat and Maren Bennewitz and Wolfram
  Burgard},
  title = {Learning Efficient Policies for Vision-based Navigation},
  booktitle = {Proc.~of the IEEE/RSJ International Conference on Intelligent 
  Robots and Systems (IROS)},
  year = {2009},
  address = {St. Louis, MO, USA},
  month = {October}
}
```
Armin Hornung.
Learning Policies for Reliable Mobile Robot Localization.
Diploma thesis, Albert-Ludwigs-Universität, Freiburg, Germany 2009.
Awarded with the VDI-Förderpreis by the Association of German Engineers
(Show abstract) (Hide abstract) (Show BibTeX) (Hide BibTeX) PDF

Cameras are a useful sensor for mobile robot localization because they are relatively cheap, compact, and lightweight. This makes them attractive for robots with payload limitations, such as humanoid robots or small unmanned aerial vehicles. However, fast movements typically reduce the ability to use a vision-based localization due to motion blur.

In this thesis, we present a reinforcement learning approach for a robot learning a vision-based navigation policy. The learned policy minimizes the time to reach the destination and implicitly takes the impact of motion blur on landmark observations and thus on localization into account. Extensive simulated and real-world experiments show that our learned policy significantly outperforms any policy of moving with a constant velocity and is generally applicable to different environments. We experimentally determined the most relevant state features for the learning task. Additionally, we present a method for compressing the learned policy with a clustering approach. While the size of the policy representation is drastically reduced, our experiments show that there is no loss of performance. This is especially valuable for memory-constrained systems.
```
@MASTERSTHESIS{hornung09diplom,
  author = {Armin Hornung},
  title = {Learning Policies for Reliable Mobile Robot Localization},
  school = {Albert-Ludwigs-Universit{\"a}t Freiburg},
  year = {2009},
  type = {Diplomarbeit},
  address = {Freiburg, Germany},
  month = {January}
}
```

2008

Dapeng Zhang and Armin Hornung.
A Table Soccer Game Recorder.
In: Video Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2008.
(Show abstract) (Hide abstract) (Show BibTeX) (Hide BibTeX)

Our table soccer robot can already challenge even professional human players. Next, the robot should play games by using human-like skills. As a foundation of this research, our table soccer game recorder can save and replay games played by humans. This video shows the construction and functionality of the recording system. We use three types of sensors mounted on a regular game table. The movement of a game rod is measured by an optical distance sensor. Its turning is observed by a magnetic rotary encoder. Two laser measurement systems are synchronized to determine the position of the ball. The raw sensor data is smoothed by an approach using multi-model Kalman filter. We developed several software modules for the system. The modules provide a basis for the future research.
```
@INPROCEEDINGS{zhang08iros,
  author = {Zhang, Dapeng and Hornung, Armin},
  title = {A Table Soccer Game Recorder},
  booktitle = {Proc.~of the IEEE/RSJ International Conference on Intelligent 
  Robots and Systems (IROS)},
  year = {2008},
  address = {Nice, France},
  month = {September}
}
```
Armin Hornung and Dapeng Zhang.
On-line Detection of Rule Violations in Table Soccer.
In KI 2008: Advances in Artificial Intelligence (KI 2008), pp. 217-224. Springer Verlag Berlin/Heidelberg; Kaiserslautern, Germany 2008.
(Show abstract) (Hide abstract) (Show BibTeX) (Hide BibTeX) PDF

In table soccer, humans can not always thoroughly observe fast actions like rod spins and kicks. However, this is necessary in order to detect rule violations for example for tournament play. We describe an automatic system using sensors on a regular soccer table to detect rule violations in realtime. Naive Bayes is used for kick classification, the parameters are trained using supervised learning. In the on-line experiments, rule violations were detected at a higher rate than by the human players. The implementation proved its usefulness by being used by humans in real games and sets a basis for future research using probability models in table soccer.
```
@INPROCEEDINGS{hornung08ki,
  author = {Hornung, Armin and Zhang, Dapeng},
  title = {On-Line Detection of Rule Violations in Table Soccer},
  booktitle = {KI '08: Proceedings of the 31st annual German conference on 
  Advances in Artificial Intelligence},
  year = {2008},
  pages = {217--224},
  address = {Berlin, Heidelberg},
  month = {March},
  publisher = {Springer-Verlag},
  doi = {http://dx.doi.org/10.1007/978-3-540-85845-4_27},
  isbn = {978-3-540-85844-7},
  location = {Kaiserslautern, Germany}
}
```
Dapeng Zhang, Bernhard Nebel and Armin Hornung.
Switching Attention Learning - A Paradigm for Introspection and Incremental Learning.
In Proceedings of the Fifth International Conference on Computational Intelligence, Robotics and Autonomous Systems (CIRAS 2008). Linz, Austria 2008.
(Show abstract) (Hide abstract) (Show BibTeX) (Hide BibTeX) PDF

Humans improve their sport skills by eliminating one recognizable weakness at a time. Inspired by this observation, we define a learning paradigm in which different learners can be plugged together. An extra attention model is in charge of iterating over them and chosing which one to improve next. The paradigm is named Switching Attention Learning (SAL). The essential idea is that improving one model in the system generates more "improvement space" for the others. Using SAL, an application for tracking the game ball in a table soccer game-recorder is implemented. We developed several models and learners which work together in the SAL framework, producing satisfying results in the experiments. The related problems, advantages, and perspective of the switching attention learning are discussed in this paper.
```
@INPROCEEDINGS{zhang08ciras,
  author = {Zhang, Dapeng and Nebel, Bernhard and Hornung, Armin},
  title = {Switching Attention Learning -- A Paradigm for Introspection and 
  Incremental Learning},
  booktitle = {Proc.~of the 5th International Conference on Computational 
  Intelligence, Robotics and Autonomous Systems (CIRAS)},
  year = {2008},
  address = {Linz, Austria},
  month = {June}
}
```

2007

Armin Hornung.
Detecting Violations in Table Soccer Games Using Naive Bayes Classifiers.
Student project, Albert-Ludwigs-Universität, Freiburg, Germany 2007.
(Show abstract) (Hide abstract)

In table soccer, humans can not always accurately observe fast actions like rod spins and kicks. However, this is necessary in order to detect rule violations for example for tournament play. This project describes an automatic referee using sensors on a regular soccer table to detect rule violations. The gap between noisy sensor data and the high-level concept of a kick is bridged by using naive Bayes classifiers for kick detection. Input to the classifiers are the coordinates of the ball relative to the kicking figure, modelled as Gaussian distributions. The classifier is trained offline using supervised learning. In the experiments, all rule violations were detected online. The implementation proved its usefulness by being used in regular games. Future work on segmenting table soccer games or sequence learning can benefit from the findings of this project by classifying game situations with the methods described here.