TechTalks from event: Technical session talks from ICRA 2012

Conference registration code to access these videos can be accessed by visiting this link: PaperPlaza. Step-by-step to access these videos are here: step-by-step process .
Why some of the videos are missing? If you had provided your consent form for your video to be published and still it is missing, please contact support@techtalks.tv

Learning and Adaptive Control of Robotic Systems I

  • RTMBA: A Real-Time Model-Based Reinforcement Learning Architecture for Robot Control Authors: Hester, Todd; Quinlan, Michael; Stone, Peter
    Reinforcement Learning (RL) is a paradigm for learning decision-making tasks that could enable robots to learn and adapt to their situation on-line. For an RL algorithm to be practical for robotic control tasks, it must learn in very few samples, while continually taking actions in real-time. Existing model-based RL methods learn in relatively few samples, but typically take too much time between each action for practical on-line learning. In this paper, we present a novel parallel architecture for model-based RL that runs in real-time by 1) taking advantage of sample-based approximate planning methods and 2) parallelizing the acting, model learning, and planning processes in a novel way such that the acting process is sufficiently fast for typical robot control cycles. We demonstrate that algorithms using this architecture perform nearly as well as methods using the typical sequential architecture when both are given unlimited time, and greatly out-perform these methods on tasks that require real-time actions such as controlling an autonomous vehicle.
  • Sensorimotor Learning of Sound Localization from an Auditory Evoked Behavior Authors: Bernard, Mathieu; PIRIM, Patrick; de Cheveigné, Alain; Gas, Bruno
    A new method for self-supervised sensorimotor learning of sound source localization is presented, that allows a simulated listener to learn online an auditorimotor map from the sensorimotor experience provided by an auditory evoked behavior. The map represents the auditory space and is used to estimate the azimuthal direction of sound sources. The learning mainly consists in non-linear dimensionality reduction of sensorimotor data. Our results show that an auditorimotor map can be learned, both from real and simulated data, and that the online learning leads to accurate estimations of azimuthal sources direction.
  • Path-following Control of a Velocity Constrained Tracked Vehicle Incorporating Adaptive Slip Estimation Authors: Burke, Michael
    This work presents a model predictive path-following controller, which incorporates adaptive slip estimation for a tracked vehicle. Tracked vehicles are capable of manoeuvring in highly variable and uneven terrain, but difficulties in their control have traditionally limited their use as autonomous platforms. Attempts to compensate for slip in environments typically require that both the forward and rotational velocities of a platform be determined, but this can be challenging. This paper shows that it is possible to estimate vehicle traction using only a rate gyroscope, by providing a suitable adaptive least squares estimator to do so. An approach to generating slip compensating controls when platform velocity constraints are applied is also presented. The approach is controller independent, but we make use of a model predictive controller, vulnerable to the effects of model-plant mismatch, to highlight the efficacy of the proposed estimation and compensation. Path following results using a mixture model to generate feasible slip values are presented, and show a significant increase in controller performance.
  • Direct Yaw Moment Control for Four Wheel Independent Steering and Drive Vehicles Based on Centripetal Force Detection Authors: Lam, Tin Lun; Xu, Yangsheng
    In this paper, a deterministic yaw moment controller for four wheel independent steering and drive vehicles is proposed to enhance driving stability and controllability. Different to conventional methods that track a desired yaw rate, the proposed controller stabilizes a vehicle by additionally tracking the heading angle of a vehicle which is more efficient and robust. The heading angle of a vehicle is obtained by a novel method which is based on centripetal force detection. It eliminates the prerequisite knowledge of the characteristics between wheels and road surface which are time varying and difficult to be measured in real time. The proposed system only requires low cost sensing equipment such as wheel speed sensor and accelerometer that makes the system practical to be utilized. The proposed heading angle detection method can be generally applied to any kind of vehicle. The deterministic yaw moment controller is also applicable to any type of four wheel independent drive vehicles.
  • Predictive Control of Chained Systems: A Necessary Condition on the Control Horizon Authors: Courtial, Estelle; Fruchard, Matthieu; Allibert, Guillaume
    This paper deals with state feedback control of chained systems based on a Nonlinear Model Predictive Control (NMPC) strategy. Chained systems can model many common nonholonomic vehicles. We establish a relation between the degree of nonholonomy and the minimum length of the control horizon so as to make the NMPC feasible. A necessary condition on the control horizon of NMPC is given and theoretically proved whatever the dimension of the chained system consid- ered. This relation is used to design a NMPC-based control strategy for chained systems. One of the advantages of NMPC is the capability of taking into account the constraints on state and on control variables. The theoretical results are illustrated through simulations on a (2,5) chained system, describing a car-like vehicle with one trailer. Difficult motion objectives that require a lateral displacement are considered.
  • Xbots: An Approach to Generating and Executing Optimal Multi-Robot Plans with Cross-Schedule Dependencies Authors: Korsah, G. Ayorkor; Kannan, Balajee; Browning, Brett; Stentz, Anthony; Dias, M. Bernardine
    In this paper, we present an approach to bounded optimal planning and flexible execution for a robot team performing a set of spatially distributed tasks related by temporal ordering constraints such as precedence or synchronization. Furthermore, the manner in which the temporal constraints are satisfied impacts the overall utility of the team, due to the existence of both routing and delay costs. We present a bounded optimal offline planner for task allocation and scheduling in the presence of such cross-schedule dependencies, and a flexible, distributed online plan execution strategy. The integrated system performs task allocation and scheduling, executes the plans smoothly in the face of real-world variations in operation speed and task execution time, and ensures graceful degradation in the event of task failure. We demonstrate the capabilities of our approach on a team of three pioneer robots operating in an indoor environment. Experimental results demonstrate that approach is effective for constrained planning and execution in the face of real-world variations.

Marine Robotics I

  • Towards Improving Mission Execution for Autonomous Gliders with an Ocean Model and Kalman Filter Authors: Smith, Ryan N.; Kelly, Jonathan; Sukhatme, Gaurav
    Effective execution of a planned path by an underwater vehicle is important for proper analysis of the gathered science data, as well as to ensure the safety of the vehicle during the mission. Here, we propose the use of an unscented Kalman filter to aid in determining how the planned mission is executed. Given a set of waypoints that define a planned path and a dicretization of the ocean currents from a regional ocean model, we present an approach to determine the time interval at which the glider should surface to maintain a prescribed tracking error, while also limiting its time on the ocean surface. We assume practical mission parameters provided from previous field trials for the problem set up, and provide the simulated results of the Kalman filter mission planning approach. The results are initially compared to data from prior field experiments in which an autonomous glider executed the same path without pre-planning. Then, the results are validated through field trials with multiple autonomous gliders implementing different surfacing intervals simultaneously while following the same path.
  • Position and Velocity Filters for Intervention AUVs Based on Single Range and Depth Measurements Authors: Viegas, Daniel; Batista, Pedro; Oliveira, Paulo; Silvestre, Carlos
    This paper proposes novel cooperative navigation solutions for an Intervention Autonomous Underwater Vehicle (I-AUV) working in tandem with an Autonomous Surface Craft (ASC). The I-AUV is assumed to be moving in the presence of constant unknown ocean currents, and aims to estimate its position relying on measurements of its range to the ASC and of its depth relatively to the sea level. Two different scenarios are considered: in one, the ASC transmits its position and velocity to the I-AUV, while in the other the ASC transmits only its position, and the I-AUV has access to measurements of its velocity relative to the ASC. A sufficient condition for observability and a method for designing state observers with Globally Asymptotically Stable (GAS) error dynamics are presented for both problems. Finally, simulation results are included and discussed to assess the performance of the proposed solutions in the presence of measurement noise.
  • Uncertainty-Driven View Planning for Underwater Inspection Authors: Hollinger, Geoffrey; Englot, Brendan; Hover, Franz; Mitra, Urbashi; Sukhatme, Gaurav
    We discuss the problem of inspecting an underwater structure, such as a submerged ship hull, with an autonomous underwater vehicle (AUV). In such scenarios, the goal is to construct an accurate 3D model of the structure and to detect any anomalies (e.g., foreign objects or deformations). We propose a method for constructing 3D meshes from sonar-derived point clouds that provides watertight surfaces, and we introduce uncertainty modeling through non-parametric Bayesian regression. Uncertainty modeling provides novel cost functions for planning the path of the AUV to minimize a metric of inspection performance. We draw connections between the resulting cost functions and submodular optimization, which provides insight into the formal properties of active perception problems. In addition, we present experimental trials that utilize profiling sonar data from ship hull inspection.
  • Formation Control of Underactuated Autonomous Surface Vessels Using Redundant Manipulator Analogs Authors: Bishop, Bradley
    In this paper, we present a method utilizing redundant manipulator analogs for formation control of underactuated autonomous surface vessels (ASVs) with realistic turning constraints and dynamics. The method used relies on casting the swarm as a single entity and utilizing redundant manipulator techniques to guarantee task-level formation control as well as obstacle avoidance and secondary tasks such as mean position control. The method presented differs from other approaches in that the units herein represent a larger class of ASVs with realistic limitations on vessel motions and that the exact position of each of the units on the formation profile is not specified.
  • Delayed State Information Filter for USBL-Aided AUV Navigation Authors: Ribas, David; Ridao, Pere; Mallios, Angelos; Palomeras, Narcis
    This paper presents a navigation system for an Autonomous Underwater Vehicle (AUV) which merges standard dead reckoning navigation data with absolute position fixes from an Ultra-Short Base Line (USBL) system. Traditionally, the USBL transceiver is located on the surface, which makes necessary to feed the position fixes back to the AUV by means of an acoustic modem. An Information filter, which maintains a bounded circular buffer of past vehicle poses, is in charge of the sensor data fusion while dealing with de delays induced by the acoustic communication. The method is validated using a data set gathered for a dam inspection task.
  • Miniature Underwater Glider: Design, Modeling, and Experimental Results Authors: Zhang, Feitian; Thon, John; Thon, Cody; Tan, Xiaobo
    The concept of gliding robotic fish combines gliding and fin-actuation mechanisms to realize energy-efficient locomotion and high maneuverability, and holds strong promise for mobile sensing in versatile aquatic environments. In this paper we present the modeling and design of a miniature fish-like glider, a key enabling component for gliding robotic fish. The full dynamics of the glider is first derived and then reduced to the sagittal plane, where the lift, drag, and pitch moment coefficients are obtained as linear or quadratic functions of the attack angle based on computational fluid dynamics (CFD) analysis. The model is used to design the glider by accommodating stringent constraints on dimensions yet meeting the desired specification on speed. A fully untethered prototype of underwater glider is developed, with a weight of 4 kg and length of 40 cm. With a net buoyancy of 20 g, it realizes a steady gliding speed of 20 cm/s. The volume and net buoyancy of this glider are less than 10% and 5%, respectively, of those of reported gliders in the literature, and its speed per unit net buoyancy is over 9 times of those other vehicles. Experimental results have shown that the model is able to capture well both the steady glide behavior under different control inputs, and the dynamics during transients.

Autonomy and Vision for UAVs

  • Cooperative Vision-Aided Inertial Navigation Using Overlapping Views Authors: Melnyk, Igor; Hesch, Joel; Roumeliotis, Stergios
    In this paper, we study the problem of Cooperative Localization (CL) for two robots, each equipped with an Inertial Measurement Unit (IMU) and a camera. We present an algorithm that enables the robots to exploit common features, observed over a sliding-window time horizon, in order to improve the localization accuracy of both robots. In contrast to existing CL methods, which require distance and/or bearing robot-to-robot observations, our algorithm infers the relative position and orientation (pose) of the robots using only the visual observations of common features in the scene. Moreover, we analyze the system observability properties to determine how many degrees of freedom (d.o.f.) of the relative transformation can be computed under different measurement scenarios. Lastly, we present simulation results to evaluate the performance of the proposed method.
  • UAV Vision: Feature Based Accurate Ground Target Localization through Propagated Initializations and Interframe Homographies Authors: Han, Kyuseo; Aeschliman, Chad; Park, Johnny; Kak, Avinash; Kwon, Hyukseong; Pack, Daniel
    Our work presents solutions to two related vexing problems in feature-based localization of ground targets in Unmanned Aerial Vehicle(UAV) images: (i) A good initial guess at the pose estimate that would speed up the convergence to the final pose estimate for each image frame in a video sequence; and (ii) Time-bounded estimation of the position of the ground target. We address both these problems within the framework of the ICP (Iterative Closest Point) algorithm that now has a rich tradition of usage in computer vision and robotics applications. We solve the first of the two problems by frame-to-frame propagation of the computed pose estimates for the purpose of the initializations needed by ICP. The second problem is solved by terminating the iterative estimation process at the expiration of the available time for each image frame. We show that when frame-to-frame homography is factored into the iterative calculations, the accuracy of the position calculated at the time of bailing out of the iterations is nearly always sufficient for the goals of UAV vision.
  • First Results in Autonomous Landing and Obstacle Avoidance by a Full-Scale Helicopter Authors: Scherer, Sebastian; Chamberlain, Lyle; Singh, Sanjiv
    Currently deployed unmanned rotorcraft rely on carefully preplanned missions and operate from prepared sites and thus avoid the need to perceive and react to the environment. Here we consider the problems of finding suitable but previously unmapped landing sites given general coordinates of the goal and planning collision free trajectories in real time to land at the “optimal” site. This requires accurate mapping, fast landing zone evaluation algorithms, and motion planning. We report here on the sensing, perception and motion planning integrated onto a full-scale helicopter that flies completely autonomously. We show results from 8 experiments for landing site selection and 5 runs at obstacles. These experiments have demonstrated the first autonomous full-scale helicopter that successfully selects its own landing sites and avoids obstacles.
  • Real-Time Onboard Visual-Inertial State Estimation and Self-Calibration of MAVs in Unknown Environments Authors: Weiss, Stephan; Achtelik, Markus W.; Lynen, Simon; Chli, Margarita; Siegwart, Roland
    The combination of visual and inertial sensors has proved to be very popular in MAV navigation due the flexibility in weight, power consumption and low cost it offers. At the same time, coping with the big latency between inertial and visual measurements and processing images in real-time impose great research challenges. Most modern MAV navigation systems avoid to explicitly tackle this by employing a ground station for off-board processing. We propose a navigation algorithm for MAVs equipped with a single camera and an IMU which is able to run onboard and in real-time. The main focus is on the proposed speed-estimation module which converts the camera into a metric body-speed sensor using IMU data within an EKF framework. We show how this module can be used for full self-calibration of the sensor suite in real-time. The module is then used both during initialization and as a fall-back solution at tracking failures of a keyframe-based VSLAM module. The latter is based on an existing high-performance algorithm, extended such that it achieves scalable 6DoF pose estimation at constant complexity. Fast onboard speed control is ensured by sole reliance on the optical flow of at least two features in two consecutive camera frames and the corresponding IMU readings. Our nonlinear observability analysis and our real experiments demonstrate that this approach can be used to control a MAV in speed, while we also show results of operation at 40 Hz on an onboard Atom computer 1.6 GHz.
  • Autonomous Landing of a VTOL UAV on a Moving Platform Using Image-Based Visual Servoing Authors: Lee, Daewon; Ryan, Tyler; Kim, H. Jin
    In this paper we describe a vision-based algorithm to control a vertical-takeoff-and-landing unmanned aerial vehicle while tracking and landing on a moving platform. Specifically, we use image-based visual servoing (IBVS) to track the platform in two-dimensional image space and generate a velocity reference command used as the input to an adaptive sliding mode controller. Compared with other vision-based control algorithms that reconstruct a full three-dimensional representation of the target, which requires precise depth estimation, IBVS is computationally cheaper since it is less sensitive to the depth estimation allowing for a faster method to obtain this estimate. To enhance velocity tracking of the sliding mode controller, an adaptive rule is described to account for the ground effect experienced during the maneuver. Finally, the IBVS algorithm integrated with the adaptive sliding mode controller for tracking and landing is validated in an experimental setup using a quadrotor.