This article addresses the challenges and opportunities of instrumenting physical world with pervasive networks sensor-rich, embedded computation. The authors present a taxonomy emerging systems outline enabling technological developments.

We consider the problem of self-deployment a mobile sensor network. are interested in deployment strategy that maximizes area coverage network with constraint each nodes has at least K neighbors, where is user-specified parameter. propose an algorithm based on artificial potential fields which distributed, scalable and does not require prior map environment. Simulations establish resulting networks have required degree high probability, well connected achieve good coverage. present...

Visual and inertial sensors, in combination, are able to provide accurate motion estimates well suited for use many robot navigation tasks. However, correct data fusion, hence overall performance, depends on careful calibration of the rigid body transform between sensors. Obtaining this information is typically difficult time-consuming, normally requires additional equipment. In paper we describe an algorithm, based unscented Kalman filter, self-calibration a camera measurement unit (IMU)....

We present the design and implementation of a real-time, vision-based landing algorithm for an autonomous helicopter. The is integrated with algorithms visual acquisition target (a helipad) navigation to target, from arbitrary initial position orientation. use vision precise detection recognition, combination Global Positioning System navigation. helicopter updates its parameters based on uses onboard behavior-based controller follow path site. significant results flight trials in field...

Successful distributed sensing and control require data to flow effectively between sensors, processors actuators on single robots, in groups across the Internet. We propose a mechanism for achieving this that we have found be powerful easy use; call it Player. Player combines an efficient message protocol with simple device model. It is implemented as multithreaded TCP socket server provides transparent network access collection of sensors actuators, often comprising robot. The abstraction...

We define a system architecture for large swarm of miniature quadcopters flying in dense formation indoors. The number small vehicles motivates novel design choices state estimation and communication. For estimation, we develop method to reliably track many rigid bodies with identical motion-capture marker arrangements. Our communication infrastructure uses compressed one-way data flow supports per radio. achieve reliable flight accurate tracking (<; 2 cm mean position error) by implementing...

We propose three sampling-based motion planning algorithms for generating informative mobile robot trajectories. The goal is to find a trajectory that maximizes an information quality metric (e.g. variance reduction, gain, or mutual information) and also falls within pre-specified budget constraint fuel, energy, time). Prior have employed combinatorial optimization techniques solve these problems, but existing are typically restricted discrete domains often scale poorly in the size of...

Recent approaches in robotics follow the insight that perception is facilitated by interaction with environment. These are subsumed under term of Interactive Perception (IP). It provides following benefits: (i) environment creates a rich sensory signal would otherwise not be present and (ii) knowledge regularity combined space data action parameters facilitate prediction interpretation signal. In this survey we postulate as principle collect evidence support analyzing categorizing existing...

We describe a method for multirobot trajectory planning in known, obstacle-rich environments. demonstrate our approach on quadrotor swarm navigating warehouse setting. Our consists of following three stages: 1) roadmap generation that generates sparse roadmaps annotated with possible interrobot collisions; 2) discrete finds valid execution schedules time and space; 3) continuous refinement creates smooth trajectories. account the downwash effect quadrotors, allowing safe flight dense...

We introduce and evaluate contact-based techniques to estimate tactile properties detect manipulation events using a biomimetic sensor. In particular, we finger forces, classify slip events. addition, present grip force controller that uses the estimation results gently pick up objects of various weights texture. The are experimentally evaluated on robotic system consisting Barrett arms hands. Our indicate able accurately forces acting in all directions, incipient slip, with over 80% success rate.

This paper introduces Robomote, a robotic solution developed to explore problems in large-scale distributed robotics and sensor networks. The design explicitly aims at enabling research networking, adhoc massively robotics, extended longevity. platform must meet many demanding criteria not limited but including: miniature size, low power, cost, simple fabrication, sensor/actuator suite that facilitates navigation localization. We argue robot test bed such as Robomote is necessary for...

We study the feasibility of extending lifetime a wireless sensor network by exploiting mobility. In our system, small percentage nodes are autonomously mobile, allowing them to move in search energy, recharge, and delivery energy immobile, energy-depleted nodes. term this approach harvesting. characterize problem uneven consumption, suggest harvesting as possible solution, provide simple analytical framework evaluate consumption scheme. Data from initial experiments using show promising results.

We present the design and implementation of a real-time, vision-based landing algorithm for an autonomous helicopter. The helicopter is required to navigate from initial position final in partially known environment based on GPS vision, locate target (a helipad shape) land it. use vision precise detection recognition. updates its parameters uses on-board behavior-based controller follow path site. results flight trials field which demonstrate that our detection, recognition control...

We describe the design and experimental validation of a large heterogeneous mobile robot team built for DARPA Software Distributed Robotics (SDR) program. The core challenge SDR program was to develop multi-robot system capable carrying out specific mission: deploy number robots into an unexplored building, map building interior, detect track intruders, transmit all above information remote operator. To satisfy these requirements, we developed consisting approximately 80 robots. sketch key...

We describe an algorithm for robot navigation using a sensor network embedded in the environment. Sensor nodes act as signposts to follow, thus obviating need map or localization on part of robot. Navigation directions are computed within (not robot) value iteration. Using small low-power radios, communicates with locally, and makes decisions based which node it is near. An processing radio signal strength data was developed so could successfully decide neighborhood belonged to. Extensive...

We describe a sensor network deployment method using autonomous flying robots. Such networks are suitable for tasks such as large-scale environmental monitoring or command and control in emergency situations. detail the algorithms used measuring connectivity provide experimental data we collected from field trials. A particular focus is on determining gaps of deployed generating plan second, repair, pass to complete connectivity. This project result collaboration between three robotics labs...

Severe energy limitations, and a paucity of computation pose set difficult design challenges for sensor networks. Recent progress in two seemingly disparate research areas namely, distributed robotics low power embedded systems has led to the creation mobile (or robotic) Autonomous node mobility brings with it its own challenges, but also alleviates some traditional problems associated static We illustrate this by presenting robomote, robot platform that functions as single network. briefly...

This paper describes a method for localizing the members of mobile robot team, using only robots themselves as landmarks, that is, we describe whereby each can determine relative range, bearing and orientation every other in without use GPS, external or instrumentation environment. Our assumes is able to measure pose nearby robots, together with changes its own pose. Using combination maximum likelihood estimation numerical optimization, subsequently infer team. basic formalism, practical...

Severe energy limitations, and a paucity of computation pose set difficult design challenges for sensor networks. Recent progress in two seemingly disparate research areas namely, distributed robotics low power embedded systems has led to the creation mobile (or robotic) Autonomous node mobility brings with it its own challenges, but also alleviates some traditional problems associated static We illustrate this by presenting robomote, robot platform that functions as single network. briefly...

We present a novel vision-based technique for navigating an unmanned aerial vehicle (UAV) through urban canyons. Our relies on both optic flow and stereo vision information. show that the combination of (stereo flow) is more effective at canyons than either alone. Optic from pair sideways looking cameras used to stay centered in canyon initiate turns junctions, while forward facing head avoid obstacles front. The was tested full autonomous tractor CSIRO part USC helicopter. Experimental...