Dynamic soaring (DS) is an aerobatic maneuver whereby a gliding aircraft harnesses energy from horizontal wind that varies in strength and/or direction to support flight. Typical approaches to dynamic soaring in autonomous unmanned aerial vehicles (UAVs) use nonlinear optimizers to generate energy-gaining trajectories, which are then followed using traditional controllers. The effectiveness of such a strategy is limited by both the local optimality of the generated trajectory, as well as controller tracking errors. In this paper, we investigate a reinforcement learning (RL) approach working in continuous space to control a DS aircraft flying in shear wind conditions. The RL controller operates in two stages: In the first stage, it observes a traditional sample-based controller flying a locally optimal DS trajectory generated a priori. In the second stage, the sample-based controller is removed and authority is passed to the RL algorithm. We show that by deviating from the original planned trajectory, the RL controller is able to achieve better performance than its baseline teacher controller.

Recent studies show that autistic children tend to speak and interact more in the presence of an interactive robot. Unfortunately, most of the robotic experiments were conducted in highly controlled clinical settings or limited selected home environments due to the high deployment cost. In this paper, we design a low-cost interactive robot that can be readily deployed to home environments. Our robot, called Lehigh Instrument for Learning Interaction (LILI), interacts with users via gestures, voice commands, and an animated speaking avatar. LILI can recognize users' faces, and her motion can be controlled either via gesture or voice. We describe both the system and software architectures of LILI. Experimental results are also presented.

A method for distributed parameter estimation of a previously unknown wind field is described. The application is dynamic soaring for small unmanned air vehicles, which severely constrains available computing while simultaneously requiring updates that are fast compared with a typical dynamic soaring cycle. A polynomial parameterization of the wind field is used, allowing implementation of a linear Kalman filter for parameter estimation. Results of Monte Carlo simulations show the effectiveness of the approach. In addition, in-flight measurements of wind speeds are compared with data obtained from video tracking of balloon launches to assess the accuracy of wind field estimates obtained using commercial autopilot modules.

In this paper, we examine the problem of extrinsic calibration of multiple LIDARs on a mobile vehicle platform. To achieve fully automated and on-line calibration, the original non-linear calibration model is reformulated as a second-order cone program (SOCP). This provides an advantage over more standard linearized approaches in that a priori information such as a default LIDAR calibration, calibration tolerances, etc., can be readily modeled. Furthermore, in contrast to general non-linear methods, the SOCP relaxation is convex, returns a global minimum, and can be solved very quickly using modern interior point methods (IPM). This enables the calibration to be estimated on-line for multiple LIDARs simultaneously. Experimental results are provided where the approach is used to successfully calibrate a pair of Sick LMS291-S14 LIDARs mounted on a mobile vehicle platform. These showed the SOCP formulation yielded a more accurate reconstruction and was 1-2 orders of magnitude faster than the traditional non-linear least-squares approach.

In this paper, we present an optimization framework for target tracking with mobile robot teams. The target tracking problem is modeled as a generic semidefinite program (SDP). When paired with an appropriate objective function, the solution to the resulting problem instance yields an optimal robot configuration for target tracking at each time-step, while guaranteeing target coverage (each target is tracked by at least one robot) and maintaining network connectivity. Our methodology is based on the graph theoretic result where the second smallest eigenvalue of the interconnection graph Laplacian matrix is a measure for the connectivity of the graph. This formulation enables us to model agent-target coverage and inter-agent communication constraints as linearmatrix inequalities.We also show that when the communication constraints can be relaxed, the resulting problem can be reposed as a second-order cone program (SOCP) which can be solved significantly more efficiently than its SDP counterpart. Simulation results for a team of robots tracking multiple targets are presented.

This paper describes an approach to cooperative localization which finds its roots in robust estimation, employing an unknown-but-bounded error model for sensor measurements. In this framework, range and bearing measurements obtained by the robots are viewed as constraints which implicitly define a set of feasible solutions in the joint configuration space of the robot team. The scheme produces bounded uncertainty estimates for the relative configuration of the team by using convex optimization techniques to approximate the projection of this feasible set onto various subspaces of the configuration space. The scheme can also be used to localize distributed sensor nodes. An important advantage of the proposed approach is that it is able to produce bounded uncertainty estimates for the relative configuration of the robots even in the case where the relative orientations of the robots are completely unknown. This is an important practical advance since errors in relative orientation are often a major contributor to positioning uncertainty in multi-robot localization schemes.

This paper investigates convex optimization strategies for coordinating a large-scale team of fully actuated mobile robots. Our primary motivation is both algorithm scalability as well as real-time performance. To accomplish this, we employ a formal definition from shape analysis for formation representation and repose the motion planning problem to one of changing (or maintaining) the shape of the formation. We then show that optimal solutions, minimizing either the total distance or minimax distance the nodes must travel, can be achieved through second-order cone programming techniques. We further prove a theoretical complexity for the shape problem of O(m1.5) as well as O(m) complexity in practice, where m denotes the number of robots in the shape configuration. Solutions for large-scale teams (1000's of robots) can be calculated in real time on a standard desktop PC. Extensions integrating both workspace and vehicle motion constraints are also presented with similar complexity bounds. We expect these results can be generalized for additional motion planning tasks, and will prove useful for improving the performance and extending the mission lives of large-scale robot formations as well as mobile ad hoc networks.

In this paper, we present a vision-based control approach for autonomously docking a wheelchair onto a vehicle lift platform. This is a principle component of the automated transport and retrieval system (ATRS) - an alternate mobility solution for drivers with lower body disabilities. The ATRS employs robotics, automation, and machine vision technologies, and can be integrated into a standard minivan or sport utility vehicle (SUV). At the core of the ATRS is a "smart" wheelchair system that autonomously navigates between the driver's position and a powered lift at the rear of the vehicle eliminating the need for an attendant. From an automation perspective, autonomously docking the wheelchair onto the lift platform presented the most significant technical challenge for the proof-of-concept ATRS. This was driven primarily by geometry constraints, which limited clearance between the chair wheels and the lift platform rails. We present significant simulation and experimental results for our approach. These indicate that the coupling of vision-based localization for feedback and input/output feedback linearization techniques for controller design can provide for accurate wheelchair navigation and reliable docking under a range of ambient illumination conditions

In this paper, we consider the task of repositioning a formation of robots to a new shape while minimizing either the maximum distance that any robot travels, or the total distance traveled by the formation. We show that optimal solutions in SE (2) can be achieved for either metric through second-order cone programming (SOCP) techniques. For the case where the orientation of the new formation shape is fixed, we obtain optimal solutions in both R2and R3. The latter also allows for complete regulation of the formation size via constraints on the shape scale. We expect that these results will prove useful for extending the mission lives of robot formations and mobile ad-hoc networks (MANETs).

The hybrid free-space optics/radio frequency (FSO/RF) network paradigm promises new levels of throughput for sensor and mobile ad-hoc networks. However, several challenges must be addressed before such a network model can be realized. These include a means by which deployed robots can autonomously establish optical links over sufficiently long distances as well as the formulation of a mobile network architecture that can exploit high throughput FSO channels. In this paper, we offer solutions to these problems in the form of hierarchical link acquisition and routing protocols. The heart of our link acquisition system (LAS) is a vision-based alignment phase for locating robot link partners via high zoom camera systems. Identification is accomplished in real-time using a multi-resolution image representation and normalized intensity distribution (NID) as a similarity metric. Our routing protocol relies upon the hierarchical state routing (HSR) model adapted to the FSO/RF paradigm. Experimental results from a large set of link acquisition trials as well as a small scale FSO/RF deployment are provided to support our approach.

In this paper, we investigate data fusion techniques for target tracking using distributed sensors. Specifically, we are interested in how pairs of bearing or range sensors can be best assigned to targets in order to minimize the expected error in the estimates. We refer to this as the focus of attention (FOA) problem. In its general form, FOA is NP-hard and not well approximable. However, for specific geometries we obtain significant approximation results: a 2-approximation algorithm for stereo cameras on a line, a PTAS for when the cameras are equidistant, and a 1.42 approximation for equally spaced range sensors on a circle. By reposing as a maximization problem - where the goal is to maximize the number of tracks with bounded error - we are able to leverage results from maximum set-packing to render the problem approximable. We demonstrate the results in simulation for a target tracking task, and for localizing a team of mobile agents in a sensor network. These results provide insights into sensor/target assignment strategies, as well as sensor placement in a distributed network.

We offer a new approach to the multi-robot localization problem. Using an unknown-but-bounded model for sensor error, we are able to define convex polytopes in the configuration space of the robot team that represent the set of configurations consistent with all sensor measurements. Estimates for the uncertainty in various parameters of the team's configuration such as the absolute position of a single robot, or the relative positions of two or more nodes can be obtained by projecting this polytope onto appropriately chosen subspaces of the configuration space. We propose a novel approach to approximating these projections using linear programming techniques. The approach can handle both bearing and range measurements with a computational complexity scaling polynomially in the number of roots. Finally, the workload is readily distributed - requiring only the communication of sensor measurements between robots. We provide simulation results for this approach implemented on a multi-robot team.

This paper presents an approach to the problem of controlling the configuration of a team of mobile agents equipped with cameras so as to optimize the quality of the estimates derived from their measurements. The issue of optimizing the robots' configuration is particularly important in the context of teams equipped with vision sensors since most estimation schemes of interest will involve some form of triangulation. We provide a theoretical framework for tackling the sensor planning problem and a practical computational strategy, inspired by work on particle filtering, for implementing the approach. We extend our previous work by showing how modeled system dynamics and configuration space obstacles can be handled. These ideas have been demonstrated both in simulation and on actual robotic platforms. The results indicate that the framework is able to solve fairly difficult sensor planning problems online without requiring excessive amounts of computational resources.

We describe a framework for cooperative control of a group of nonholonomic mobile robots that allows us to build complex systems from simple controllers and estimators. The resultant modular approach is attractive because of the potential for reusability. Our approach to composition also guarantees stability and convergence in a wide range of tasks. There are two key features in our approach: 1) a paradigm for switching between simple decentralized controllers that allows for changes in formation; 2) the use of information from a single type of sensor, an omnidirectional camera, for all our controllers. We describe estimators that abstract the sensory information at different levels, enabling both decentralized and centralized cooperative control. Our results include numerical simulations and experiments using a testbed consisting of three nonholonomic robots.

The paper presents an approach to the problem of controlling the configuration of a team of mobile agents equipped with cameras so as to optimize the quality of the estimates derived from their measurements. The issue of optimizing the robots' configuration is particularly important in the context of teams equipped with vision sensors since most estimation schemes of interest will involve some form of triangulation. We provide a theoretical framework for tackling the sensor planning problem and a practical computational strategy, inspired by work on particle filtering, for implementing the approach. The ideas have been demonstrated both in simulation and on actual robotic platforms. The results indicate that the framework is able to solve fairly difficult sensor planning problems online without requiring excessive amounts of computational resources.

This paper considers the problem of vision-based control of a nonholonomic mobile robot. We describe the design and implementation of real-time estimation and control algorithms on a car-like robot platform using a single omni-directional camera as a sensor without explicit use of odometry. We provide experimental results for each of these vision-based control objects. The algorithms are packaged as control modes and can be combined hierarchically to perform higher level tasks involving multiple robots.

We describe a framework for coordinating multiple robots in cooperative manipulation tasks in which vision is used for establishing relative position and orientation and maintaining formation. The two key contributions are a cooperative scheme for localizing the robots based on visual imagery that is more robust than decentralized localization, and a set of control algorithms that allow the robots to maintain a prescribed formation (shape and size). The ability to maintain a prescribed formation allows the robots to "trap" objects in their midst, and to "flow" the formation to a desired position. We derive the cooperative localization and control algorithms and present experimental results that illustrate the implementation and the performance of these algorithms.