ShareCam is a robotic pan, tilt, and zoom webbased camera controlled by simultaneous frame requests from online users. Part I describes the system. This paper, Part II, focuses on algorithms. The ShareCam problem is to find a camera frame that optimizes a measure of total user satisfaction. We present a grid-based approximation algorithm: given camera frame requests from n users, and approximation bound #, we analyze the tradeoff between solution quality and processing speed and prove that the algorithm runs in O(n/# ) time. The algorithm can be distributed to run in O(1/# ) time at each client and in O(n + 1/# ) time at the server. Experiments suggest that performance of the distributed algorithm degrades gracefully as clients fail to complete their part of the computation. ShareCam can be found online at: http://www.tele-actor.net/sharecam/.

ShareCam is a robotic pan, tilt, and zoom web-based camera controlled by simultaneous frame requests from online users. Part II describes algorithms. This paper, Part I, focuses on the system. Robotic webcameras are commercially available but currently restrict control to only one user at a time. ShareCam introduces a new interface that allows simultaneous control by many users. In this Java-based interface, participating users interact from their remotely located browsers where users draw desired frames over a fixed panoramic image. User inputs are transmitted back to a pair of PC servers that compute optimal camera parameters, servo the camera, and provide a video stream to all users. We describe the system, online experiments, and compare results with two frame selection models based on user ``satisfaction,'' one memoryless and the second based on satisfaction over multiple motion cycles. ShareCam is available online at: www.tele-actor.net/sharecam/

We propose the ShareCam Problem: controlling a single robotic pan, tilt, zoom camera based on simultaneous frame requests from n online users. To solve it, we propose a new piecewise linear metric, Intersection Over Maximum (IOM), for the degree of satisfaction for each users. To maximize overall satisfaction, we present several algorithms. For a discrete set of m distinct zoom levels, we give an exact algorithm that runs in O(n m) time. The algorithm can be distributed to run in O(nm) time at each client and in O(n log n + mn) time at the server.

As a new application area for Automation, Near Real Time Satellite Imaging provides timely optical information which is used for weather prediction, disaster control, surveillance, and military applications. As the satellite passes over the earth, camera imaging parameters are changed to focus and capture data within requested zones. The Satellite Frame Selection problem arises when there are competing client requests: we want to automatically choose the satellite frame parameters that will maximize "reward" during each time window. In this paper we propose a new reward metric that incorporates both image resolution and coverage. For a set of n client requests we give a series of algorithms, the fastest computes optimal results in O(n ) for satellites with continuously variable resolution. We implement the algorithms and compare computation speed for all algorithms.

The scientific field study of wildlife often requires vigilant observation of detailed animal behavior over extended periods. In remote and inhospitable locations, observation can be an arduous, expensive, and dangerous experience for field scientists. We are developing a new class of networked teleoperated robotic “observatories ” that allows “citizen scientists ” and professional scientists to remotely observe, record, and index animal activity and behaviors via the internet. This paper describes CONE-Welder, installed at the Rob & Bessie Welder Wildlife Foundation in Texas to gather photographic and quantitative data for a biological study of avian activity and hypothesized range change for selected subtropical bird species. Since the system was deployed on 12 May 2008, over 600 users (“players”) have participated online. Players have requested over 2.2 million camera frames and captured over 29,000 photographs. Within these photos, citizen scientists have classified 74 unique species, including eight avian species previously unknown to have breeding populations within the region. The collected dataset quantifies seasonal presence of birds of particular interest, e.g., the Green Jay (Cyanocorax incas). This paper describes the system architecture, the game interface that provides incentives for player participation, and initial data collected. CONE-Welder is available online at:

Systems and Algorithms for Collaborative Teleoperation Collaborative Teleoperation (CT) systems allow many users to simultaneously share control of a single remote physical resource such as a robot or human "explorer", with applications in education, health care, journalism, and entertainment. A primary challenge is scalable methods for computing consensus commands. This thesis combines results from two networked CT systems: one with a robotic webcamera (the Co-Opticon) and one with a human explorer (the Tele-Actor).

Abstract — When constructing buildings, frequent inspection and detailed visual documentation are important but may not be feasible in remote or dangerous environments. We describe a networked robotic camera system that can automatically monitor construction details and allow remote human experts to zoom in on features as construction proceeds to archive the construction process over time, thereby reducing travel cost and human risk. We describe system architecture, interface design, data structures, and algorithms for such systems. We also report initial experimental results from cameras at two outdoor construction sites. Index Terms — automation, construction, networked robots, pan-tilt-zoom camera, panoramic display. I.

Abstract — A new class of low-cost teleoperated pan-tilt-zoom robotic video cameras can provide high resolution panoramic displays of remote sites for disaster response, environmental monitoring, and security applications. While the camera is teleoperated, the resulting video is transmitted back and inserted into an evolving panoramic display. Since small errors in camera position can produce large registration errors in the panoramic display, we address the image alignment problem. To quantify alignment error, we introduce a new metric based on motor error and image overlap. We use this metric to develop a fast minimal variance image alignment algorithm. We have implemented the algorithm and describe experiments demonstrating panoramic quality and that optimal alignment can be computed as fast as the camera can be tele-operated. Index Terms — tele-operation, telerobotics, networked robot, panoramic display, pan-tilt-zoom camera.

Collaborative Frame Selection arises when one robotic pan, tilt, zoom camera is shared by many users. The problem is to compute optimal camera parameters based on simultaneous frame requests from all users. We formalize the problem using a new metric, Intersection Over Maximum (IOM), to model the degree of satisfaction for each user, and seek to maximize total satisfaction for n users. We assume the zoom parameter is chosen from a discrete set of m levels and consider cases with discrete and continuous pan and tilt paramters. For a discrete set of w h pan and tilt values, we give an exact algorithm that runs in O((n + mwh) log n). For continuous pan and tilt, we give an exact algorithm that runs in O(n m) time. We also give a distributed version that runs in O(nm) time at each client and in O(n log n+mn) time at the server. An implementation of the second algorithm can be found online at: http://www.tele-actor.net/sharecam/.