Showing papers in "IEEE Robotics & Automation Magazine in 2010"
TL;DR: In the last five years, advances in materials, electronics, sensors, and batteries havefueled a growth in the development of microunmanned aerial vehicles (MAVs) that are between 0.1 and 0.5 m in length and0.1-0.5 kg in mass.
Abstract: In the last five years, advances in materials, electronics, sensors, and batteries have fueled a growth in the development of microunmanned aerial vehicles (MAVs) that are between 0.1 and 0.5 m in length and 0.1-0.5 kg in mass [1]. A few groups have built and analyzed MAVs in the 10-cm range [2], [3]. One of the smallest MAV is the Picoftyer with a 60-mmpropellor diameter and a mass of 3.3 g [4]. Platforms in the 50-cm range are more prevalent with several groups having built and flown systems of this size [5]-[7]. In fact, there are severalcommercially available radiocontrolled (PvC) helicopters and research-grade helicopters in this size range [8].
806 citations
TL;DR: An approach based on HMM, GMR, and dynamical systems to allow robots to acquire new skills by imitation was presented and evaluated and applications on different kinds of robots were presented to highlight the flexibility of the proposed approach.
Abstract: We presented and evaluated an approach based on HMM, GMR, and dynamical systems to allow robots to acquire new skills by imitation. Using HMM allowed us to get rid of the explicit time dependency that was considered in our previous work [12], by encapsulating precedence information within the statistical representation. In the context of separated learning and reproduction processes, this novel formulation was systematically evaluated with respect to our previous approach, LWR [20], LWPR [21], and DMPs [13]. We finally presented applications on different kinds of robots to highlight the flexibility of the proposed approach in three different learning by imitation scenarios.
448 citations
204 citations
TL;DR: This article summarizes the discussion and findings from the "Teaching Humans About Human-Robot Interaction" workshop on the development of an HRI course for computer scientists and engineers.
Abstract: The field of human-robot interaction (HRI) addresses the design, understanding, and evaluation of robotic systems, which involve humans and robots interacting through communication. As the field matures, education of students becomes increasingly important. Courses in HRI provide the canonical set of knowledge and core skills that represent the current state of the field and permit the evolution of knowledge and methods to be transferred from research to a broad set of students. In addition, coursework in HRI creates a workforce capable of transferring HRI theory to practice. However, as would be expected with an emerging field, HRI courses are largely ad hoc. This article summarizes the discussion and findings from the "Teaching Humans About Human-Robot Interaction" workshop on the development of an HRI course for computer scientists and engineers.
194 citations
TL;DR: In this paper, the authors extend the capabilities of the harmonic potential field (HPF) approach to cover both the kinematic and dynamic aspects of a robot's motion by augmenting it with a novel type of damping forces.
Abstract: This article extends the capabilities of the harmonic potential field (HPF) approach to planning to cover both the kinematic and dynamic aspects of a robot's motion. The suggested approach converts the gradient guidance field from a harmonic potential to a control signal by augmenting it with a novel type of damping forces called nonlinear, anisotropic, damping forces (NADF). The HPF (harmonic potential field) approach to planning is emerging as a powerful paradigm for the guidance of autonomous agents.
177 citations
TL;DR: It is shown that imitation learning has helped significantly to start learning with reasonable initial behavior, however, many applications are still restricted to rather lowdimensional domains and toy applications.
Abstract: Recent trends in robot learning are to use trajectory-based optimal control techniques and reinforcement learning to scale complex robotic systems. On the one hand, increased computational power and multiprocessing, and on the other hand, probabilistic reinforcement learning methods and function approximation, have contributed to a steadily increasing interest in robot learning. Imitation learning has helped significantly to start learning with reasonable initial behavior. However, many applications are still restricted to rather lowdimensional domains and toy applications. Future work will have to demonstrate the continual and autonomous learning abilities, which were alluded to in the introduction.
172 citations
TL;DR: The role of software components as architectural units of large, possibly distributed, software-intensive robotic systems and on design techniques to assemble components into systems are discussed.
Abstract: This article discusses the role of software components as architectural units of large, possibly distributed, software-intensive robotic systems. The focus is on technologies to manage the heterogeneity of hardware, computational, and communication resources and on design techniques to assemble components into systems .A component-based system is a composition of components, and the way components interact with other components and with the computational environment greatly affects the flexibility of the entire system and the reusability of individual functionality.
150 citations
TL;DR: This article describes the dynamical system MPs representation in a way that it is straightforward to reproduce, and presents an appropriate imitation learning method, i.e., locally weighted regression, which can be used both for initializing RL tasks as well as for modifying the start-up phase in a rhythmic task.
Abstract: In this article, we present both novel learning algorithms and experiments using the dynamical system MPs. As such, we describe this MP representation in a way that it is straightforward to reproduce. We review an appropriate imitation learning method, i.e., locally weighted regression, and show how this method can be used both for initializing RL tasks as well as for modifying the start-up phase in a rhythmic task. We also show our current best-suited RL algorithm for this framework, i.e., PoWER. We present two complex motor tasks, i.e., ball-in-a-cup and ball paddling, learned on a real, physical Barrett WAM, using the methods presented in this article. Of particular interest is the ball-paddling application, as it requires a combination of both rhythmic and discrete dynamical systems MPs during the start-up phase to achieve a particular task.
135 citations
TL;DR: The paper describes application areas, societal drivers, motivating scenarios, desired system capabilities, and fundamental research areas that should be considered in the design of medical and health-care robots.
Abstract: The aim of this article is to propose some of the most important capabilities and technical achievements of medical and health-care robotics needed to improve human health and well-being. The paper describes application areas, societal drivers, motivating scenarios, desired system capabilities, and fundamental research areas that should be considered in the design of medical and health-care robots.
133 citations
TL;DR: A detailed retrospective on modular robots is presented and connections between modular robots and programmable matter are discussed, using distributed algorithms that use a modules ability to observe its current neighborhood and local rules to decide what to do next.
Abstract: We have presented a detailed retrospective on modular robots and discussed connections between modular robots and programmable matter. This field has seen a great deal of creativity and innovation at the level of designing physical systems capable of matching shape to function and algorithms that achieve this capability. The success of these projects rests on the convergence of innovation in hardware design and materials for creating the basic building blocks, information distribution for programming the interaction between the blocks, and control. Most current systems have dimensions on the order of centimeters, yet pack computation, communication, sensing, and power transfer capabilities into their form factors. Additionally, these modules operate using distributed algorithms that use a modules ability to observe its current neighborhood and local rules to decide what to do next.
117 citations
TL;DR: Using methods from operations research to address a fundamental routing problem involving heterogeneous UAVs and using the well-known Lin-Kernighan-Helsgaun heuristic was applied to the transformed ATSP.
Abstract: Heterogeneous unmanned aerial vehicles (UAVs) are being developed for several civil and military applications. These vehicles can differ either in their motion constraints or sensing/attack capabilities. This article uses methods from operations research to address a fundamental routing problem involving heterogeneous UAVs. The approach is to transform the routing problem into a relatively better understood single, asymmetric, traveling salesman problem (ATSP) and use the algorithms available for the ATSP to address the routing problem. To test the effectiveness of the transformation, the well-known Lin-Kernighan-Helsgaun heuristic was applied to the transformed ATSP. Computational results on the transformed ATSP show that solutions whose costs are within 16% of the optimum can be obtained relatively fast [within 40 s of central processing unit (CPU)] for the routing problem involving ten heterogeneous UAVs and 40 targets.
TL;DR: The autonomous system, Crusher, developed for the UPI program is described and the learning approaches that aided in its successful performance are described.
Abstract: Autonomous navigation by a mobile robot through L natural, unstructured terrain is one of the premier k challenges in field robotics. Tremendous advances V in autonomous navigation have been made recently in field robotics. Machine learning has played an increasingly important role in these advances. The Defense Advanced Research Projects Agency (DARPA) UGCV-Perceptor Integration (UPI) program was conceived to take a fresh approach to all aspects of autonomous outdoor mobile robot design, from vehicle design to the design of perception and control systems with the goal of achieving a leap in performance to enable the next generation of robotic applications in commercial, industrial, and military applications. The essential problem addressed by the UPI program is to enable safe autonomous traverse of a robot from Point A to Point B in the least time possible given a series of waypoints in complex, unstructured terrain separated by 0.2-2 km. To accomplish this goal, machine learning techniques were heavily used to provide robust and adaptive performance, while simultaneously reducing the required development and deployment time. This article describes the autonomous system, Crusher, developed for the UPI program and the learning approaches that aided in its successful performance.
TL;DR: In this article, the authors focus on the PR2, a robot running the Robot Operating System (ROS) for free for two years through the Willow Garage's PR2 Beta Program.
Abstract: In the last issue of IEEE Robotics & Automation Magazine, we talked about a number of different robots running the robot operating system (ROS). In this issue, we are going to focus on one in particular, the PR2. Willow Garage announced in May the 11 institutions that will get to use a PR2 for free for two years through its PR2 Beta Program (Figure 1). The PR2s all run ROS, and the institutions have all committed to making significant open-source contributions back to the ROS community. ROS has been gaining traction at an astonishing rate since its release in January, but having 11 PR2s at top institutions will accelerate that progress even further. We will give a brief overview of what the PR2 Beta Program recipients are planning to do, and then dive into a concrete example of the use of ROS on the PR2.
TL;DR: The paper describes the current evolution of the class of Fo¿laga underwater vehicles, whose actuation mechanism is a hybrid between oceanographic gliders and standard, self-propelled autonomous underwater vehicles (AUVs).
Abstract: The paper describes the current evolution of the class of Fo?laga underwater vehicles, whose actuation mechanism is a hybrid between oceanographic gliders and standard, self-propelled autonomous underwater vehicles (AUVs). The Fo?laga vehicles have been designed for coastal oceanography and environmental sampling; the application-oriented design approach has resulted in light-weight, low-cost, low-maintenance vehicles. A description of the vehicles' design and guidance, navigation and control capabilities is given, together with data from recent experimental trials.
TL;DR: Connecting all of these components, which form an aquatic robotic system, in synchronous operation can greatly assist the scientists in improving the overall understanding of the complex ocean environment.
Abstract: More than 70% of our earth is covered by water, yet we have explored less than 5% of the aquatic environment. Aquatic robots, such as autonomous underwater vehicles (AUVs), and their supporting infrastructure play a major role in the collection of oceanographic data. To make new discoveries and improve our overall understanding of the ocean, scientists must make use of these platforms by implementing effective monitoring and sampling techniques to study ocean upwelling, tidal mixing, and other ocean processes. Effective observation and continual monitoring of a dynamic system as complex as the ocean cannot be done with one instrument in a fixed location. A more practical approach is to deploy a collection of static and mobile sensors, where the information gleaned from the acquired data is distributed across the network. Additionally, orchestrating a multisensor, long-term deployment with a high volume of distributed data involves a robust, rapid, and cost-effective communication network. Connecting all of these components, which form an aquatic robotic system, in synchronous operation can greatly assist the scientists in improving our overall understanding of the complex ocean environment.
Journal Article•
TL;DR: This tutorial highlights the differences between solving a rigid-body problem using 3-D vectors and solving the same problem using spatial vectors, so that the reader can get an idea of what it means to think in 6-D.
Abstract: The purpose of this tutorial is to present a beginner's guide to 6-D vectors in sufficient detail that a reader can begin to use them as a practical problem-solving tool right away. This tutorial covers the basics, and Part 2 will cover the application of 6-D vectors to a variety of robot kinematics and dynamics calculations. The rest of the tutorial is chiefly concerned with explaining what spatial vectors are and how to use them. It highlights the differences between solving a rigid-body problem using 3-D vectors and solving the same problem using spatial vectors, so that the reader can get an idea of what it means to think in 6-D.
TL;DR: The robot demonstrates that this SMA actuator is power efficient compared with other SMA designs that cannot operate with batteries, and it is showed that a rotational version of this type of actuators is comparable with an electromagnetic motor.
Abstract: We have presented a simple but effective method to design flexible actuators. This process relies on understanding the behavior of a simple unit cell element built out of SMA sheet. The unit cell effectively uses the properties of flat SMA sheets: it operates in the bent region where more force is generated; it minimizes the nonbent SMA; and it heats up only the bent regions. However, the force generated by this unit cell does not scale up well, and an array of them is needed to increase the force generated. Building an actuator out of an array of unit cells increases its complexity but provides advantages, including the control of expansion length, trajectory, and generated force. Given the current technologies, including 3-D printing and laser cutting, a variety of support structures can be built to create an actuator with a given behavior. We have shown three types of configurations: linear, rotational, and surface. The linear actuator has been tested for endurance and can easily perform over 10,000 repetitions under load without breaking. These actuators have been tested in actual systems, such as the battery-operated HexRoller robot that uses six actuators connected in a chain. The robot demonstrates that this SMA actuator is power efficient compared with other SMA designs that cannot operate with batteries. We have also showed that a rotational version of this type of actuator is comparable with an electromagnetic motor.
TL;DR: This column is designed to introduce you and help track this important community effort and to help decide whether or not to download and try the ROS system.
Abstract: Robot operating system (ROS) is a free and opensource system that has grown out of a novel collaboration between industry and academia. This column is designed to introduce you and help track this important community effort. The latest information about ROS will always be available on the Web (http://ros.org). My goal is to help you decide whether or not to download and try the system. In future columns, I'll write about the latest developments in ROS and its progress in the ROS community.
Journal Article•
TL;DR: The role of software components as architectural units of large, possibly distributed, softwareintensive robotic systems and the challenges that make the development of reusable components and flexible component systems difficult in robotics are discussed.
Abstract: This article is the second of a two-part series intended as an introduction to Component-based Software Engineering (CBSE) in Robotics. In Part I, we regarded a component as a piece of software that implements robotic functionality. The focus was on design principles and implementation guidelines that enable the development of reusable and maintainable software building blocks. In Part II, we discuss the role of software components as architectural units of large, possibly distributed, softwareintensive robotic systems. The focus is on technologies to manage the heterogeneity of hardware, computational, and communication resources and on design techniques to assemble components into systems. A Component-based System is a composition of components and the way components interact with other components and with the computational environment greatly affects the flexibility of the entire system and the reusability of individual functionality. The paper is structured as follows. Section II discusses the challenges that make the development of reusable components and flexible component systems difficult in robotics. In particular, we analyze the different sources of variability in robotic technology and applications and the corresponding requirements for flexibility in robotic software component-based systems. The subsequent four sections introduce key concepts related to the development of reusable software components, such as threading, synchronization, resource awareness, distribution, and quality of service negotiation and illustrate architectural models that enable the independent evolution of different variability concerns of a component-based robotic system. They are classified in four categories related to main concerns of component-based systems, namely Computation (Section III), Configuration (Section IV), Communication (Section V), and Coordination (Section VI). Finally, Section VII draws the relevant conclusions. It should be noted that this paper is not a survey of the state-of-the-art in component-based software engineering, nor in robotics software development. While there is an increasing
Journal Article•
TL;DR: The goal is to acquire the Earth-referenced Cartesian coordinates of a known target, with the necessary accuracy required for positioning the vehicle, so that the target falls within the manipulator workspace.
Abstract: This contribution focuses on the problem, intrinsic to autonomous underwater manipulation, of medium-range target localization for guiding the vehicle toward the target area. Based on the use of the dual-frequency identification sonar (DIDSON) sonar, the goal is to acquire the Earth-referenced Cartesian coordinates of a known target, with the necessary accuracy required for positioning the vehicle, so that the target falls within the manipulator workspace.
TL;DR: This tutorial shall examine the application of spatial vectors to various problems in robot kinematics and dynamics and consider both the mathematical solution of a problem and the computer code to calculate the answer.
Abstract: In this tutorial, we shall examine the application of spatial vectors to various problems in robot kinematics and dynamics. To demonstrate that spatial vectors are both a tool for analysis and a tool for computation, we shall consider both the mathematical solution of a problem and the computer code to calculate the answer.
TL;DR: In this article, it is considered how a very large numbers of robots may be made to coexist, communicate, and compete fairly toward achieving their individual goals, i.e., to build a society of I robots.
Abstract: In this article, we consider how a very large numbers of robots, differing in their bodies, sensing, and intelligence, may be made to coexist, communicate, and compete fairly toward achieving their individual goals, i.e., to build a society of I robots. We discuss some characteristics that the rules defining acceptable social behaviors should possess. We consider threats that may be posed to such a society by the misbehaviors of some of its members, either due to faults or malice, and the possibility to detect and isolate them through cooperation of peers. The article presents examples of motion control protocols, for arbitrarily large groups of heterogeneous robots. We discuss intrusion detection algorithms, which allow detection of deviance from such rules, and algorithms to build a consensus view on the environment and on the integrity of peers, so as to improve the overall security of the society of robots.
[...]
TL;DR: This article proposes auction aggregation protocols (AAPs) for task assignment in multihop wireless robot networks by proposing traversal aggregation algorithm [routing with face traversal (RFT)-routing toward the event with the traversal of the face containing the event] and is an optimal solution.
Abstract: This article proposes auction aggregation protocols (AAPs) for task assignment in multihop wireless robot networks. A robot collector leads an auction and initiates a response tree construction by transmitting the search message. After receiving the message, each robot makes a decision on whether to retransmit a search message, based on the estimated response cost of its robots, up to fe-hops away. Robots wait to receive the bids from its children in the search tree. Then, robots aggregate responses by selecting the best bid and forward it back toward the robot collector (auctioning robot). When distance is used as the sole cost metrics, the traversal aggregation algorithm [routing with face traversal (RFT)-routing toward the event with the traversal of the face containing the event] can be applied and is an optimal solution. Several other protocols and their enhancements are also described here.
Journal Article•
TL;DR: The vocabulary of security protocol analysis is applied to this new environment, considering how changes in technology call into question assumptions of trust that seem natural to users of video teleconferencing and robot telepresence.
Abstract: Recent advances in affect recognition, facial expression synthesis, and speech synthesis are giving nonexperts access to tools for performing live manipulation of audio and video streams, as well as control signals for telepresence robots. We apply the vocabulary of security protocol analysis to this new environment, considering how changes in technology call into question assumptions of trust that seem natural to users of video teleconferencing and robot telepresence.
Journal Article•
Journal Article•
[...]
TL;DR: A cognitive reference model of architectures and behaviors of cognitive robots is presented, which reveals the architectural differences and behavioral characteristics of Cognitive robots beyond conventional imperative robots.
Abstract: To model the cognitive processes of robot authentication, an emerging technology of cognitive robots is introduced. A number of fundamental problem remains in cognitive robot studies, such as: 1) What are the necessary and sufficient intelligent behaviors of cognitive robots? and 2) How the intelligent capabilities of cognitive robots are distinguished from those of their imperative counterparts? This article presents a cognitive reference model of architectures and behaviors of cognitive robots, which reveals the architectural differences and behavioral characteristics of cognitive robots beyond conventional imperative robots. Cognitive informatics (CI) foundations of cognitive robots are explored from the aspects of neural informatics (Nel) and abstract intelligence (αl). The architectural model of cognitive robots is described based on a layered reference model of the brain (LRMB). The behavioral model of cognitive robot is elaborated with the generic behavioral model and the hierarchical relations among the behavioral processes of cognitive robots. A reference model of cognitive robots (RMCRs) is derived, which models cognitive robots by seven forms of intelligent behaviors at the imperative, autonomic, and cognitive layers from the bottom-up. Applications of RMCR are described in robot authentication, computational intelligence (Col), and automation systems.
TL;DR: In the "Industrial Activities" column of the September 2009 issue of IEEE Robotics & Automation Magazine, recent world-wide road-mapping activities were compared with a focus such that the current European initiatives could only very briefly be described as mentioned in this paper.
Abstract: In the "Industrial Activities" column of the September 2009 issue of IEEE Robotics & Automation Magazine, recent world-wide road-mapping activities were compared with a focus such that the current European initiatives could only very briefly be described. This column is geared toward describing these activities in more detail.