Refine
Year of publication
Institute
Has Fulltext
- no (136) (remove)
Document Type
- Article (71)
- Conference Proceeding (56)
- Part of a Book (7)
- Book (2)
Keywords
- Autonomous mobile robots (2)
- Industry 4.0 (2)
- Multi-robot systems (2)
- Smart factory (2)
- Anomaly detection (1)
- Automation (1)
- Benchmark (1)
- Computational modeling (1)
- Control (1)
- Cyber-physical systems (1)
Robots are widely used as a vehicle to spark interest in science and technology in learners. A number of initiatives focus on this issue, for instance, the Roberta Initiative, the FIRST Lego League, the World Robot Olympiad and RoboCup Junior. Robotic competitions are valuable not only for school learners but also for university students, as the RoboCup initiative shows. Besides technical skills, the students get some project exposure and experience what it means to finish their tasks on time. But qualifying students for future high-tech areas should not only be for students from developed countries. In this article, we present our experiences with research and education in robotics within the RoboCup initiative, in Germany and South Africa; we report on our experiences with trying to get the RoboCup initiative in South Africa going. RoboCup has a huge support base of academic institutions in Germany; this is not the case in South Africa. We present our ‘north–south’ collaboration initiatives in RoboCup between Germany and South Africa and discuss some of the reasons why we think it is harder to run RoboCup in South Africa.
The field of Cognitive Robotics aims at intelligent decision making of autonomous robots. It has matured over the last 25 or so years quite a bit. That is, a number of high-level control languages and architectures have emerged from the field. One concern in this regard is the action language GOLOG. GOLOG has been used in a rather large number of applications as a high-level control language ranging from intelligent service robots to soccer robots. For the lower level robot software, the Robot Operating System (ROS) has been around for more than a decade now and it has developed into the standard middleware for robot applications. ROS provides a large number of packages for standard tasks in robotics like localisation, navigation, and object recognition. Interestingly enough, only little work within ROS has gone into the high-level control of robots. In this paper, we describe our approach to marry the GOLOG action language with ROS. In particular, we present our architecture on inte grating golog++, which is based on the GOLOG dialect Readylog, with the Robot Operating System. With an example application on the Pepper service robot, we show how primitive actions can be easily mapped to the ROS ActionLib framework and present our control architecture in detail.
The Carologistics team participates in the RoboCup Logistics League for the seventh year. The RCLL requires precise vision,
manipulation and path planning, as well as complex high-level decision
making and multi-robot coordination. We outline our approach with an
emphasis on recent modifications to those components.
The team members in 2018 are David Bosen, Christoph Gollok, Mostafa
Gomaa, Daniel Habering, Till Hofmann, Nicolas Limpert, Sebastian Schönitz,
Morian Sonnet, Carsten Stoffels, and Tarik Viehmann.
This paper is based on the last year’s team description.
Autonomous agents require rich environment models for fulfilling their missions. High-definition maps are a well-established map format which allows for representing semantic information besides the usual geometric information of the environment. These are, for instance, road shapes, road markings, traffic signs or barriers. The geometric resolution of HD maps can be as precise as of centimetre level. In this paper, we report on our approach of using HD maps as a map representation for autonomous load-haul-dump vehicles in open-pit mining operations. As the mine undergoes constant change, we also need to constantly update the map. Therefore, we follow a lifelong mapping approach for updating the HD maps based on camera-based object detection and GPS data. We show our mapping algorithm based on the Lanelet 2 map format and show our integration with the navigation stack of the Robot Operating System. We present experimental results on our lifelong mapping approach from a real open-pit mine.
In this paper we report on CO2 Meter, a do-it-yourself carbon dioxide measuring device for the classroom. Part of the current measures for dealing with the SARS-CoV-2 pandemic is proper ventilation in indoor settings. This is especially important in schools with students coming back to the classroom even with high incidents rates. Static ventilation patterns do not consider the individual situation for a particular class. Influencing factors like the type of activity, the physical structure or the room occupancy are not incorporated. Also, existing devices are rather expensive and often provide only limited information and only locally without any networking. This leaves the potential of analysing the situation across different settings untapped. Carbon dioxide level can be used as an indicator of air quality, in general, and of aerosol load in particular. Since, according to the latest findings, SARS-CoV-2 can be transmitted primarily in the form of aerosols, carbon dioxide may be used as a proxy for the risk of a virus infection. Hence, schools could improve the indoor air quality and potentially reduce the infection risk if they actually had measuring devices available in the classroom. Our device supports schools in ventilation and it allows for collecting data over the Internet to enable a detailed data analysis and model generation. First deployments in schools at different levels were received very positively. A pilot installation with a larger data collection and analysis is underway.
One central challenge for self-driving cars is a proper path-planning. Once a trajectory has been found, the next challenge is to accurately and safely follow the precalculated path. The model-predictive controller (MPC) is a common approach for the lateral control of autonomous vehicles. The MPC uses a vehicle dynamics model to predict the future states of the vehicle for a given prediction horizon. However, in order to achieve real-time path control, the computational load is usually large, which leads to short prediction horizons. To deal with the computational load, the control algorithm can be parallelized on the graphics processing unit (GPU). In contrast to the widely used stochastic methods, in this paper we propose a deterministic approach based on grid search. Our approach focuses on systematically discovering the search area with different levels of granularity. To achieve this, we split the optimization algorithm into multiple iterations. The best sequence of each iteration is then used as an initial solution to the next iteration. The granularity increases, resulting in smooth and predictable steering angle sequences. We present a novel GPU-based algorithm and show its accuracy and realtime abilities with a number of real-world experiments.
In this paper we present an extension of the action language Golog that allows for using fuzzy notions in non-deterministic argument choices and the reward function in decision-theoretic planning. Often, in decision-theoretic planning, it is cumbersome to specify the set of values to pick from in the non-deterministic-choice-of-argument statement. Also, even for domain experts, it is not always easy to specify a reward function. Instead of providing a finite domain for values in the non-deterministic-choice-of-argument statement in Golog, we now allow for stating the argument domain by simply providing a formula over linguistic terms and fuzzy uents. In Golog’s forward-search DT planning algorithm, these formulas are evaluated in order to find the agent’s optimal policy. We illustrate this in the Diner Domain where the agent needs to calculate the optimal serving order.
The RoboCup Logistics League (RCLL) is a robotics competition in a production logistics scenario in the context of a Smart Factory. In the competition, a team of three robots needs to assemble products to fulfill various orders that are requested online during the game. This year, the Carologistics team was able to win the competition with a new approach to multi-agent coordination as well as significant changes to the robot’s perception unit and a pragmatic network setup using the cellular network instead of WiFi. In this paper, we describe the major components of our approach with a focus on the changes compared to the last physical competition in 2019.
In the future, we expect manufacturing companies to follow a new paradigm that mandates more automation and autonomy in production processes. Such smart factories will offer a variety of production technologies as services that can be combined ad hoc to produce a large number of different product types and variants cost-effectively even in small lot sizes. This is enabled by cyber-physical systems that feature flexible automated planning methods for production scheduling, execution control, and in-factory logistics.
During development, testbeds are required to determine the applicability of integrated systems in such scenarios. Furthermore, benchmarks are needed to quantify and compare system performance in these industry-inspired scenarios at a comprehensible and manageable size which is, at the same time, complex enough to yield meaningful results.
In this chapter, based on our experience in the RoboCup Logistics League (RCLL) as a specific example, we derive a generic blueprint for how a holistic benchmark can be developed, which combines a specific scenario with a set of key performance indicators as metrics to evaluate the overall integrated system and its components.
Rugged terrain robot designs are important for field robotics missions. A number of commercial platforms are available, however, at an impressive price. In this paper, we describe the hardware and software component of a low-cost wheeled rugged-terrain robot. The robot is based on an electric children quad bike and is modified to be driven by wire. In terms of climbing properties, operation time and payload it can compete with some of the commercially available platforms, but at a far lower price.
The work in modern open-pit and underground mines requires the transportation of large amounts of resources between fixed points. The navigation to these fixed points is a repetitive task that can be automated. The challenge in automating the navigation of vehicles commonly used in mines is the systemic properties of such vehicles. Many mining vehicles, such as the one we have used in the research for this paper, use steering systems with an articulated joint bending the vehicle’s drive axis to change its course and a hydraulic drive system to actuate axial drive components or the movements of tippers if available. To address the difficulties of controlling such a vehicle, we present a model-predictive approach for controlling the vehicle. While the control optimisation based on a parallel error minimisation of the predicted state has already been established in the past, we provide insight into the design and implementation of an MPC for an articulated mining vehicle and show the results of real-world experiments in an open-pit mine environment.
The Robot Operating System (ROS) is the current de-facto standard in robot middlewares. The steadily increasing size of the user base results in a greater demand for training as well. User groups range from students in academia to industry professionals with a broad spectrum of developers in between. To deliver high quality training and education to any of these audiences, educators need to tailor individual curricula for any such training. In this paper, we present an approach to ease compiling curricula for ROS trainings based on a taxonomy of the teaching contents. The instructor can select a set of dedicated learning units and the system will automatically compile the teaching material based on the dependencies of the units selected and a set of parameters for a particular training. We walk through an example training to illustrate our work.
In this work, we report on our attempt to design and implement an early introduction to basic robotics principles for children at kindergarten age. One of the main challenges of this effort is to explain complex robotics contents in a way that pre-school children could follow the basic principles and ideas using examples from their world of experience. What sets apart our effort from other work is that part of the lecturing is actually done by a robot itself and that a quiz at the end of the lesson is done using robots as well. The humanoid robot Pepper from Softbank, which is a great platform for human–robot interaction experiments, was used to present a lecture on robotics by reading out the contents to the children making use of its speech synthesis capability. A quiz in a Runaround-game-show style after the lecture activated the children to recap the contents they acquired about how mobile robots work in principle. In this quiz, two LEGO Mindstorm EV3 robots were used to implement a strongly interactive scenario. Besides the thrill of being exposed to a mobile robot that would also react to the children, they were very excited and at the same time very concentrated. We got very positive feedback from the children as well as from their educators. To the best of our knowledge, this is one of only few attempts to use a robot like Pepper not as a tele-teaching tool, but as the teacher itself in order to engage pre-school children with complex robotics contents.
This paper presents an approach for reducing the cognitive load for humans working in quality control (QC) for production processes that adhere to the 6σ -methodology. While 100% QC requires every part to be inspected, this task can be reduced when a human-in-the-loop QC process gets supported by an anomaly detection system that only presents those parts for manual inspection that have a significant likelihood of being defective. This approach shows good results when applied to image-based QC for metal textile products.
Logic-based robot control in highly dynamic domains / Ferrein, Alexander ; Lakemeyer, Gerhard
(2008)
In this paper we report on an architecture for a self-driving car that is based on ROS2. Self-driving cars have to take decisions based on their sensory input in real-time, providing high reliability with a strong demand in functional safety. In principle, self-driving cars are robots. However, typical robot software, in general, and the previous version of the Robot Operating System (ROS), in particular, does not always meet these requirements. With the successor ROS2 the situation has changed and it might be considered as a solution for automated and autonomous driving. Existing robotic software based on ROS was not ready for safety critical applications like self-driving cars. We propose an architecture for using ROS2 for a self-driving car that enables safe and reliable real-time behaviour, but keeping the advantages of ROS such as a distributed architecture and standardised message types. First experiments with an automated real passenger car at lower and higher speed-levels show that our approach seems feasible for autonomous driving under the necessary real-time conditions.
Hybrid control for autonomous systems — Integrating learning, deliberation and reactive control
(2010)
Reasoning with Qualitative Positional Information for Domestic Domains in the Situation Calculus
(2011)
Cyber-physical systems are ever more common in manufacturing industries. Increasing their autonomy has been declared an explicit goal, for example, as part of the Industry 4.0 vision. To achieve this system intelligence, principled and software-driven methods are required to analyze sensing data, make goal-directed decisions, and eventually execute and monitor chosen tasks. In this chapter, we present a number of knowledge-based approaches to these problems and case studies with in-depth evaluation results of several different implementations for groups of autonomous mobile robots performing in-house logistics in a smart factory. We focus on knowledge-based systems because besides providing expressive languages and capable reasoning techniques, they also allow for explaining how a particular sequence of actions came about, for example, in the case of a failure.
In this paper we present CAESAR, an intelligent domestic service robot. In domestic settings for service robots complex tasks have to be accomplished. Those tasks benefit from deliberation, from robust action execution and from flexible methods for human–robot interaction that account for qualitative notions used in natural language as well as human fallibility. Our robot CAESAR deploys AI techniques on several levels of its system architecture. On the low-level side, system modules for localization or navigation make, for instance, use of path-planning methods, heuristic search, and Bayesian filters. For face recognition and human–machine interaction, random trees and well-known methods from natural language processing are deployed. For deliberation, we use the robot programming and plan language READYLOG, which was developed for the high-level control of agents and robots; it allows combining programming the behaviour using planning to find a course of action. READYLOG is a variant of the robot programming language Golog. We extended READYLOG to be able to cope with qualitative notions of space frequently used by humans, such as “near” and “far”. This facilitates human–robot interaction by bridging the gap between human natural language and the numerical values needed by the robot. Further, we use READYLOG to increase the flexible interpretation of human commands with decision-theoretic planning. We give an overview of the different methods deployed in CAESAR and show the applicability of a system equipped with these AI techniques in domestic service robotics
Many biped robots deploy a form of gait that follows the zero moment point (ZMP) approach, that is, the robot is in a stable position at any point in time. This requires the robot to be fully actuated. While very stable, the draw-backs of this approach are a fairly slow gait and high energy consumption. An alternative approach is the so-called passive-dynamic walking, where the gait makes use of the inertia and dynamic stability of the robot. In this paper we describe our ongoing work of combining the principles of passive-dynamic walking on the fully-actuated biped robot Nao, which is also deployed for robotic soccer applications. We present a simple controller that allows the robot to stably rock sidewards, showing a closed limit-cycle. We discuss first results of superimposing a forward motion on the sidewards motion. Based on this we expect to endow the Nao with a fast, robust, and stable passive-dynamic walk on the fully-actuated Nao in the future.
Embedding fuzzy controllers in golog / Ferrein, Alexander ; Schiffer, Stefan ; Lakemeyer, Gerhard
(2009)
RGB-D sensors such as the Microsoft Kinect or the Asus Xtion are inexpensive 3D sensors. A depth image is computed by calculating the distortion of a known infrared light (IR) pattern which is projected into the scene. While these sensors are great devices they have some limitations. The distance they can measure is limited and they suffer from reflection problems on transparent, shiny, or very matte and absorbing objects. If more than one RGB-D camera is used the IR patterns interfere with each other. This results in a massive loss of depth information. In this paper, we present a simple and powerful method to overcome these problems. We propose a stereo RGB-D camera system which uses the pros of RGB-D cameras and combine them with the pros of stereo camera systems. The idea is to utilize the IR images of each two sensors as a stereo pair to generate a depth map. The IR patterns emitted by IR projectors are exploited here to enhance the dense stereo matching even if the observed objects or surfaces are texture-less or transparent. The resulting disparity map is then fused with the depth map offered by the RGB-D sensor to fill the regions and the holes that appear because of interference, or due to transparent or reflective objects. Our results show that the density of depth information is increased especially for transparent, shiny or matte objects.
South Africa in recent years is the establishment of a number of research hubs involved in AI activities ranging from mobile robotics and computational intelligence, to knowledge representation and reasoning, and human language technologies. In this survey we take the reader through a quick tour of the research being conducted at these hubs, and touch on an initiative to maintain and extend the current level of interest in AI research in the country.