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 .

Localization and Mapping

• Point Set Registration through Minimization of the L2 Distance between 3D-NDT Models Authors: Stoyanov, Todor; Magnusson, Martin; Lilienthal, Achim, J.
Point set registration --- the task of finding the best fitting alignment between two sets of point samples, is an important problem in mobile robotics. This article proposes a novel registration algorithm, based on the distance between Three-Dimensional Normal Distributions Transforms. 3D-NDT models --- a sub-class of Gaussian Mixture Models with uniformly weighted, largely disjoint components, can be quickly computed from range point data. The proposed algorithm constructs 3D-NDT representations of the input point sets and then formulates an objective function based on the \$L_2\$ distance between the considered models. Analytic first and second order derivatives of the objective function are computed and used in a standard Newton method optimization scheme, to obtain the best-fitting transformation. The proposed algorithm is evaluated and shown to be more accurate and faster, compared to a state of the art implementation of the Iterative Closest Point and 3D-NDT Point-to-Distribution algorithms.
• Consistency Analysis for Sliding-Window Visual Odometry Authors: Dong-Si, Tue-Cuong; Mourikis, Anastasios
In this paper we focus on the problem of {em visual odometry}, i.e., the task of tracking the pose of a moving platform using visual measurements. In recent years, several VO algorithms have been proposed that employ nonlinear minimization in a sliding window of poses for this task. Through the use of iterative re-linearization, these methods are capable of successfully addressing the nonlinearity of the measurement models, and have become the de-facto standard for high-precision VO. In this work, we conduct an analysis of the properties of marginalization, which is the process through which older states are removed from the sliding window. This analysis shows that the standard way of marginalizing older poses results in an erroneous change in the rank of the measurements' information matrix, and leads to underestimation of the uncertainty of the state estimates. Based on the analytical results, we also propose a simple modification of the way in which the measurement Jacobians are computed. This modification avoids the above problem, and results in an algorithm with superior accuracy, as demonstrated in both simulation tests and real-world experiments.
• Efficient Visual Odometry Using a Structure-Driven Temporal Map Authors: Martinez-Carranza, Jose; Calway, Andrew
We describe a method for visual odometry using a single camera based on an EKF framework. Previous work has shown that &#64257;ltering based approaches can achieve accuracy performance comparable to that of optimisation methods providing that large numbers of features are used. However, computational requirements are signicantly increased and frame rates are low. We address this by employing higher level structure - in the form of planes - to ef&#64257;ciently parameterise features and so reduce the &#64257;lter state size and computational load. Moreover, we extend a 1-point RANSAC outlier rejection method to the case of features lying on planes. Results of experiments with both simulated and real-world data demonstrate that the method is effective, achieving comparable accuracy whilst running at signi&#64257;cantly higher frame rates.
• Using Depth in Visual Simultaneous Localisation and Mapping Authors: Scherer, Sebastian Andreas; Dube, Daniel; Zell, Andreas
We present a method of utilizing depth information as provided by RGBD sensors for robust real-time visual simultaneous localisation and mapping (SLAM) by augmenting monocular visual SLAM to take into account depth data. This is implemented based on the freely available software â€œParallel Tracking and Mappingâ€ (PTAM) by Georg Klein, which was originally developed for augmented reality applications. Our modi&#64257;cations allow PTAM to be used as a 6D visual SLAM system even without any additional information about odometry or from an inertial measurement unit.
• A Visual Marker for Precise Pose Estimation Based on Lenticular Lenses Authors: Tanaka, Hideyuki; Sumi, Yasushi; Matsumoto, Yoshio
Visual marker is a useful assistive tool for service robots. Existing planar visual markers have poor accuracy and stability in pose estimation, especially in frontal direction. In this study, we developed a novel visual marker based on a new principle enabling accurate and stable pose estimation even by observation from frontal direction. The marker has moire patterns which consist of lenticular lens and stripe pattern, which vary their appearance according to visual-line angle of observation. We can extract pose information from the pattern by a single camera. We developed a prototype of the marker and an algorithm for pose estimation, and then demonstrated its superiority to existing markers by some validation tests.
• Robot Semantic Mapping through Wearable Sensor-Based Human Activity Recognition Authors: Sheng, Weihua; Li, Gang; Zhu, Chun; Du, Jianhao; Cheng, Qi
Semantic information can help both humans and robots to understand their environments better. In order to obtain semantic information efficiently and link it to a metric map, we present a semantic mapping approach through human activity recognition in an indoor human-robot coexisting environment. An intelligent mobile robot platform can create a 2D metric map, while human activity can be recognized using motion data from wearable motion sensors mounted on a human subject. Combined with pre-learned models of activity-to-furniture type association and robot pose estimates, the robot can determine the distribution of the furniture types on the 2D metric map. Simulations and real world experiments demonstrate that the proposed method is able to create a reliable metric map with accurate semantic information.

Climbing Robots

• Step Negotiation with Wheel Traction: A Strategy for a Wheel-Legged Robot Authors: Turker, Korhan; Sharf, Inna
This paper presents a quasi-static step climbing behavior for a minimal sensing wheel-legged quadruped robot called PAW. In the quasi-static climbing maneuver, the robot benefits from wheel traction and uses its legs to reconfigure itself with respect to the step during the climb. The control methodology with the corresponding controller parameters is determined and the state machine for the maneuver is developed. With this controller, PAW is able to climb steps higher than its body clearance. Furthermore, any step height up to this maximum achievable height can be negotiated autonomously with a single set of controller parameters, without knowledge of the step height or distance to the step.
• Fast Accessible Rescue Device by Using a Flexible Sliding Actuator Authors: Tsukagoshi, Hideyuki
This paper discusses a method of locomotion called the â€œfluid powered ropewayâ€. It aims to collect information in dangerous buildings as rapidly and safely as possible. The device is mainly composed of a flexible flat tube and a gondola probe driven by fluid power using the buckling phenomenon of the tube. The big advantage is the gondola has the potential to traverse rocky terrains that wheeled and crawler-type vehicles have difficulty in crossing over. This is because the drive force of the gondola is not against the ground but against the tube. In this paper, first, how to operate fluid powered ropeway in a disaster site is illustrated. Next, how to increase the drive force, how to enhance the ability of the gondola to travel over obstacles, and an analysis of the performance are discussed. Finally, the feasibility of the proposed method is verified through an experiment that uses the prototype developed.
• Design Considerations for Attachment and Detachment in Robot Climbing with Hot Melt Adhesives Authors: Wang, Liyu; Neuschaefer, Fabian; Bernet, Remo; Iida, Fumiya
Robust climbing in unstructured environments is a long-standing challenge in robotics research. Recently there has been an increasing interest in using adhesive materials for that purpose. For example, a climbing robot using hot melt adhesives (HMAs) has demonstrated advantages in high attachment strength, reasonable operation costs, and applicability to different surfaces. Despite the advantages, there still remain several problems related to the attachment and detachment operations, which prevent this approach from being used in a broader range of applications. Among others, one of the main problems lies in the fact that the adhesive characteristics of this material were not fully understood fin the context of robotic climbing locomotion. As a result, the previous robot often could not achieve expected locomotion performances and ``contaminated'' the environment with HMAs left behind. In order to improve the locomotion performances, this paper focuses on attachment and detachment operations in robot climbing with HMAs. By systematically analyzing the adhesive property and bonding strength of HMAs to different materials, we propose a novel detachment mechanism that substantially improves climbing performances without HMA traces.
• Parameter Optimization of Directional Dry Adhesives for Robotic Climbing and Gripping Applications Authors: Ruffatto III, Donald; Spenko, Matthew
This paper experimentally investigates the optimization of directional dry adhesives that can be used for robotic climbing and gripping applications. Directional dry adhesives are modeled on gecko setae. The adhesives are comprised of arrays of micro-scale polymer stalks. The geometry of the polymer stalks has a significant effect upon their adhesion properties. A set of parameters including stalk thickness, stalk angle, face angle and stalk curvature have been identified as factors that influence both normal and shear adhesion levels. A new micro-resolution rapid prototyping process is used to create adhesives with varying geometry and advanced features such as curved stalks. A series of experimental tests characterize the significance of each parameter. Tests indicate that the new curved stalk geometry presented here can provide the greatest overall adhesion and robustness to variations in pull-off angle.
• System and Design of Clothbot: A Robot for Flexible Clothes Climbing Authors: Liu, Yuanyuan; Wu, Xinyu; QIAN, Huihuan; Zheng, Duan; sun, jianquan; Xu, Yangsheng
This paper presents a novel climbing robot called Clothbot which has high maneuverability on flexible clothes. It has a novel gripper consisting of two parallel wheels that can grip continuously and stably on various kinds of clothes. Clothbot also has an omni-directional tail of two DOFs so that it can change its center of gravity to control the moving direction on complex and undeterminate clothes. Consequently, Clothbot is able to access most positions of the clothes by moving straight and turning around with only four motors. It is compact, small and light-weighted but has a load capacity six times its own weight. A series of experiments validate its high performance on flexible clothes.

Embodied Inteligence - iCUB

• Learning Reusable Task Components Using Hierarchical Activity Grammars with Uncertainties Authors: Lee, Kyuhwa; Kim, Tae-Kyun; Demiris, Yiannis
We present a novel learning method using activity grammars capable of learning reusable task components from a reasonably small number of samples under noisy conditions. Our linguistic approach aims to extract the hierarchical structure of activities which can be recursively applied to help recognize unforeseen, more complicated tasks that share the same underlying structures. To achieve this goal, our method 1) actively searches for frequently occurring action symbols that are subset of input samples to effectively discover the hierarchy, and 2) explicitly takes into account the uncertainty values associated with input symbols due to the noise inherent in low-level detectors. In addition to experimenting with a synthetic dataset to systematically analyze the algorithm's performance, we apply our method in human-led imitation learning environment where a robot learns reusable components of the task from short demonstrations to correctly imitate more complicated, longer demonstrations of the same task category. The results suggest that under reasonable amount of noise, our method is capable to capture the reusable structures of tasks and generalize to cope with recursions.
• Stabilization for the Compliant Humanoid Robot COMAN Exploiting Intrinsic and Controlled Compliance Authors: Li, Zhibin; Vanderborght, Bram; Tsagarakis, Nikolaos; Colasanto, Luca; Caldwell, Darwin G.
The work presents the standing stabilization of a compliant humanoid robot against external force disturbances and variations of the terrain inclination. The novel contribution is the proposed control scheme which consists of three strategies named compliance control in the transversal plane, body attitude control, and potential energy control, all combined with the intrinsic passive compliance in the robot. The physical compliant elements of the robot are exploited to react at the first instance of the impact while the active compliance control is applied to further absorb the impact and dissipate the elastic energy stored in springs preventing the high rate of spring recoil. The body attitude controller meanwhile regulates the spin angular momentum to provide more agile reactions by changing body inclination. The potential energy control module constrains the robot center of mass (COM) in a virtual slope to convert the excessive kinetic energy into potential energy to prevent falling. Experiments were carried out with the proposed balance stabilization control demonstrating superior balance performance. The compliant humanoid was capable of recovering from external force disturbances and moderate or even abrupt variations of the terrain inclination. Experimental data such as the impulse forces, real COM, center of pressure (COP) and the spring elastic energy are presented and analyzed.
• Efficient Human-Like Walking for the COmpliant Humanoid COMAN Based on Kinematic Motion Primitives (kMPs) Authors: Moro, Federico Lorenzo; Tsagarakis, Nikolaos; Caldwell, Darwin G.
Research in humanoid robotics in recent years has led to significant advances in terms of the ability to walk and even run. Yet, despite the general achievements in locomotion and control, energy efficiency is still one important area that requires further attention, especially as it is one of the major steeping stones leading to increased autonomy. This paper examines, and quantifies, the energetic benefits of introducing passive compliance into bipedal locomotion using COMAN, an intrinsically COmpliant huMANoid robot. The novelty of the method proposed consists of: i) the use of a method of gait synthesis based on kinematic Motion Primitives (kMPs) extracted from human, ii) the frequency tuning of the resultant trajectories, to excite the physical elasticity of the system, and the subsequent analysis of the energetic performance of the robot. The motivation is to assess the possible effects of using dynamic human-like, and human derived, trajectories, with significant Center of Mass (CoM) vertical displacement, regulated in frequency around the frequency band of the system resonances, on the excitation of the compliant actuators, and subsequently to measure and verify any energetic benefit. Experimental results show that if the gait frequency is close to one of the main resonant frequencies of the robot, then the total work contribution of the elastic compliant element to the overall motion of the robot is positive (15% of the work required is generated by the springs).
• Closed-loop primitives: A method to generate and recognize reaching actions from demonstration Authors: Parlaktuna, Mustafa; Tunaoglu, Doruk; Ugur, Emre; Sahin, Erol
The studies on mirror neurons observed in monkeys indicate that recognition of otherâ€™s actions activates neural circuits that are also responsible for generating the very same actions in the animal. The mirror neuron hypothesis argues that such an overlap between action generation and recognition can provide a shared worldview among individuals and be a key pillar for communication. Inspired by these findings, this paper extends a learning by demonstration method for online recognition of observed actions. The proposed method is shown to recognize and generate different reaching actions demonstrated by a human on a humanoid robot platform. Experiments show that the proposed method is robust to both occlusions during the observed actions as well as variances in the speed of the observed actions. The results are successfully demonstrated in an interactive game with the iCub humanoid robot platform.
• Active Object Recognition on a Humanoid Robot Authors: Browatzki, Bjoern; Tikhanoff, Vadim; Metta, Giorgio; Buelthoff, Heinrich H.; Wallraven, Christian
Interaction with its environment is a key requisite for a humanoid robot. Especially the ability to recognize and manipulate unknown objects is crucial to successfully work in natural environments. Visual object recognition, however, still remains a challenging problem, as three-dimensional objects often give rise to ambiguous, two-dimensional views. Here, we propose a perception-driven, multisensory exploration and recognition scheme to actively resolve ambiguities that emerge at certain viewpoints. We define an efficient method to acquire two-dimensional views in an object-centered task space and sample characteristic views on a view sphere. Information is accumulated during the recognition process and used to select actions expected to be most beneficial in discriminating similar objects. Besides visual information we take into account proprioceptive information to create more reliable hypotheses. Simulation and real-world results clearly demonstrate the efficiency of active, multisensory exploration over passive, vision-only recognition methods.
• Imitation Learning of Non-Linear Point-To-Point Robot Motions Using Dirichlet Processes Authors: Krueger, Volker; Tikhanoff, Vadim; Natale, Lorenzo; SANDINI, GIULIO
In this paper we discuss the use of the infinite Gaussian mixture model and Dirichlet processes for learning robot movements from demonstrations. Starting point of this work is an earlier paper where the authors learn a non- linear dynamic robot movement model from a small number of observations. The model in that work is learned using a classical finite Gaussian mixture model (FGMM) where the Gaussian mixtures are appropriately constrained. The problem with this approach is that one needs to make a good guess for how many mixtures the FGMM should use. In this work, we generalize this approach to use an infinite Gaussian mixture model (IGMM) which does not have this limitation. Instead, the IGMM automatically finds the number of mixtures that are necessary to reflect the data complexity. For use in the context of a non-linear dynamic model, we develop a Constrained IGMM (CIGMM). We validate our algorithm on the same data that was used in [5], where the authors use motion capture devices to record the demonstrations. As further validation we test our approach on novel data acquired on our iCub in a different demonstration scenario in which the robot is physically driven by the human demonstrator.