This paper presents the results of a number of demonstrations which illustrate the performance benefit which can be gained through the cooperation of a team of Unmanned Aerial Vehicles (UAVs). A team of Brumby MkIII. ...
详细信息
ISBN:
(纸本)9783642001956
This paper presents the results of a number of demonstrations which illustrate the performance benefit which can be gained through the cooperation of a team of Unmanned Aerial Vehicles (UAVs). A team of Brumby MkIII. UAVs performed a. mission which required estimating the location of a number of ground features given some prior information. The UAVs dynamically plan their paths sharing (1) feature. state information obtained from observations made by onboard vision sensors and (2) information based utilities which allow for globally optimal action planning. Three demonstrations were performed with varying levels of cooperation. The results are compared for each case and it is shown how, by enabling cooperation through the sharing of utilities, large performance increases can be gained.
Arthropods with their extraordinary locomotive capabilities have inspired roboticists, giving rise to major accomplishments in robotics research over the past decade. Most notably bio-inspired hexapod robots using onl...
详细信息
ISBN:
(纸本)9783642001956
Arthropods with their extraordinary locomotive capabilities have inspired roboticists, giving rise to major accomplishments in robotics research over the past decade. Most notably bio-inspired hexapod robots using only task level open-loop controllers [22, 9] exhibit stable dynamic locomotion over highly broken and unstable terrain. We present experimental data on the dynamics of SprawlHex- a hexapod robot with adjustable body sprawl - consisting of time trajectory of full body configuration and single leg ground reaction forces. The dynamics of SprawlHexis compared and contrasted to that of insects. Spraw/Hexdynamics has qualitative similarities to that of insects in both sagittal and horizontal plane. Spraw/Hexpresents a step towards construction of an effective physical model to study arthropod locomotion.
Human interactive robots continue to improve human quality of life with their diverse applications. Their field includes, but is not limited to, haptic devices, force feedback tele-manipulation, surgical co-manipulati...
详细信息
ISBN:
(纸本)9783642001956
Human interactive robots continue to improve human quality of life with their diverse applications. Their field includes, but is not limited to, haptic devices, force feedback tele-manipulation, surgical co-manipulation, medical rehabilitation, and various multi-degree of freedom robotic devices where the human operator and robot are often required to simultaneously execute tasks and collaborate with a specific share of forces/energy. More than tuning mechanical design, the robot control enhancement with a force sensor, is the key for increasing transparency (i.e the capacity for a robot to follow human movements without any human-perceptible resistive forces). With ail ideal robot control, the interaction between robot and human Would be extremely natural and fluid that the comanipulation of tasks Would seem to be achieved with a transparent aid from the robot. For Such, the classical force feedback control in certain cases still seems insufficient as is often limited by various factors (noise, bandwidth limitation, stability, sensor cost..etc). Our experiments are focused on evaluating the performance increase in terms of transparency of controller by using human motion predictions. We evaluate several ways to use predictive informations in the control to overcome present transparency limitations during a simple comanipulation pointing task.
While there has been significant progress in recent years in the study of estimation and control of dynamic network graphs, limited attention has been paid to the experimental validation and verification of such algor...
详细信息
ISBN:
(纸本)9783642001956
While there has been significant progress in recent years in the study of estimation and control of dynamic network graphs, limited attention has been paid to the experimental validation and verification of such algorithms on distributed teams of robots. In this work we conduct an experimental study of a non-trivial distributed connectivity control algorithm on a team of seven nonholonomic robots as well as in simulation. The implementation of the algorithm is completely decentralized and asynchronous, assuming that each robot only has access to its pose and knowledge of the total number of robots. All other necessary information is determined via message passing with neighboring robots. We show that such algorithms, requiring complex inter-agent communication and coordination, are feasible as well as highly successful in enabling a network of robots to adapt to disturbances while preserving connectivity.
Sensor models directly influence the efficiency quid robustness of the estimation processes used in robot and object localization. This paper focuses on a probabilistic range finder sensor model for dynamic environmen...
详细信息
ISBN:
(纸本)9783642001956
Sensor models directly influence the efficiency quid robustness of the estimation processes used in robot and object localization. This paper focuses on a probabilistic range finder sensor model for dynamic environments. The dynamic nature results from the presence of unmodeled and possibly moving objects and people. The goal of this paper is twofold. First, we present experiments to validate the Rigorously, Bayesian Beam Model (RBBM), a new model we proposed in a previous paper. Second, we propose a sample-based full scan model to improve the state of the art models. In contrast to these Gaussian-based state of the art full scan models, the proposed model is able to handle the multi-modality of the range finder data, which is shown here to occur even in simple static environments.
Introduction: Recognizing previously observed locations is a significant challenge in the areas of autonomous mapping and navigation. The place recognition problem arises in applications ranging from pose initializati...
详细信息
ISBN:
(纸本)9783642001956
Introduction: Recognizing previously observed locations is a significant challenge in the areas of autonomous mapping and navigation. The place recognition problem arises in applications ranging from pose initialization and localization using prior maps, closing loops and relocalizing lost robots in simultaneous localization and mapping (SLAM), and merging maps collected at different times or by multiple robots. In this paper, we present and evaluate a general approach for efficiently recognizing and identifying previously observed areas. While we focus on two-dimensional laser point cloud maps, the methodology described is applicable to other sensing modalities and map representations. The maps we consider are constructed from large datasets taken over tens to hundreds of kilometers under regular driving conditions and road speeds in urban and suburban environments. At this scale, efficiency-especially in the case of online SLAM, which must be solved in real-time-is critical.
As personal robots enter the social environments of our workplaces and homes, it will be important for them to be able to learn from a wide demographic of people. Our research seeks to identify simple, natural, and pr...
详细信息
ISBN:
(纸本)9783642001956
As personal robots enter the social environments of our workplaces and homes, it will be important for them to be able to learn from a wide demographic of people. Our research seeks to identify simple, natural, and prevalent human teaching cues that are useful for directing the attention of robot learners so that robots can learn efficiently and effectively from these interactions. This research goal is significant for several reasons. First, most people do not have expertise in robotics or machine learning techniques. Second, personal robots will have to learn new tasks and skills within the bounds of human attention and patience. Third, people bring a lifetime of experience in learning from and teaching others and naturally do quite a lot to socially structure appropriate learning environment's and interactions so that others can learn efficiently and effectively. Personal robots should be able to leverage theses social interactions to also learn efficiently and effectively from people. The hypothesis that we investigate in this paper concerns the structure of social behavior and embodied interaction that we term "social filters:" dynamic, embodied cues through which the teacher can guide the behavior of the robot by emphasizing and de-emphasizing objects in the environment. Specifically, we con, lecture that action timing and "spatial scaffolding" are two important kinds of social filters - simple, natural, and prevalent non-verbal cues in which teachers time their actions and use their bodies to structure the learning environment to direct the attention of the learner. The challenge is to identify when and how these cues are employed in natural learning settings, and how they can be used by a robot to efficiently guide its internal attention and learning processes.
"Robotic Mapping and Exploration" is an important contribution in the area of simultaneous localization and mapping (SLAM) for autonomous robots, which has been receiving a great deal of attention by the res...
详细信息
ISBN:
(数字)9783642010972
ISBN:
(纸本)9783642010965
"Robotic Mapping and Exploration" is an important contribution in the area of simultaneous localization and mapping (SLAM) for autonomous robots, which has been receiving a great deal of attention by the research community in the latest few years. The contents are focused on the autonomous mapping learning problem. Solutions include uncertainty-driven exploration, active loop closing, coordination of multiple robots, learning and incorporating background knowledge, and dealing with dynamic environments. Results are accompanied by a rich set of experiments, revealing a promising outlook toward the application to a wide range of mobile robots and field settings, such as search and rescue, transportation tasks, or automated vacuum cleaning.
The monograph written by Dezhen Song is focused on a robotic camera simultaneously controlled by multiple online users via the Internet. A challenging match between the collaboratively tele-operated robotic cameras an...
详细信息
ISBN:
(数字)9783540880653
ISBN:
(纸本)9783540880646
The monograph written by Dezhen Song is focused on a robotic camera simultaneously controlled by multiple online users via the Internet. A challenging match between the collaboratively tele-operated robotic cameras and the needs from nature environment observation is sought, which greatly extends the domain of online robots in both application and technology development directions, including building construction site monitoring, public space surveillance, and distance education. New solutions are proposed which demonstrate the enormous potential of Internet-based infrastructures for immediate success in the market.
This book describes the development of a robot mapping and navigation system inspired by models of the neural mechanisms underlying spatial navigation in the rodent hippocampus. Computational models of animal navigati...
ISBN:
(数字)9783540775201
ISBN:
(纸本)9783540775195
This book describes the development of a robot mapping and navigation system inspired by models of the neural mechanisms underlying spatial navigation in the rodent hippocampus. Computational models of animal navigation systems have traditionally had limited performance when implemented on robots. The aim of the work was to determine the extent to which hippocampal models can be used to provide a robot with functional mapping and navigation capabilities in real world environments. The focus of the research was on achieving practical robot performance, rather than maintaining biological plausibility.
暂无评论