ICRA 2015 Papers

Functionality of a human hand can be augmented with extra robotic fingers attached to the forearm. These wearable extra fingers can work together with the human fingers and perform tasks that are usually difficult for a single hand. Such technology may benefit amputees and surviving stroke patients who are forced to live with only one functional arm. In this paper, we present a novel method, exploiting redundancies in the elbow, to control a pair of extra robotic fingers, termed Supernumerary Robotic Fingers or SR Fingers for short, to perform “hold-and-manipulate” tasks. Two control protocols are explored here. One protocol enables discrete transition between different modes of motion, while the other continuously controls grasping behavior of the SR Fingers. Four robot assisted “hold-and-manipulate” tasks are examined, including twisting off the cap of a water bottle, opening the lid of a food container, plugging a cable into an electronic device, and mixing salad in a bowl. Through the successful completion of these tasks, we demonstrate that wearable extra fingers have the potential to provide those with impaired hands the opportunity to live with more independence and work more productively.

The controllable adhesives used by insects to both carry large loads and move quickly despite their small scale inspires the μTug robot concept. These are small robots that can both move quickly and use controllable adhesion to apply interaction forces many times their body weight. The adhesives enable these autonomous robots to accomplish this feat on a variety of common surfaces without complex infrastructure. The benefits, requirements, and theoretical efficiency of the adhesive in this application are discussed as well as the practical choices of actuator and robot working surface material selection. A robot actuated by piezoelectric bimorphs demonstrates fast walking with a no-load rate of 50 Hz and a loaded rate of 10 Hz. A 12 g shape memory alloy (SMA) actuated robot demonstrates the ability to load more of the adhesive enabling it to tow 6.5 kg on glass (or 500 times its body weight). Continuous rotation actuators (electromagnetic in this case) are demonstrated on another 12 g robot give it nearly unlimited work cycles through gearing. This leads to advantages in towing capacity (up to 22 kg or over 1800 times its body weight), step size, and efficiency. This work shows that using such an adhesive system enables small robots to provide truly human scale interaction forces, despite their size and mass. This will enable future microrobots to not only sense the state of the human environment in which they operate, but apply large enough forces to modify it in response.

Similar to the atmosphere, coherent structures, e.g., fronts, exist in the ocean. These frontal structures are known to be highly productive, supporting the whole spectrum of marine life. Ocean fronts are dynamic in time and space, and are a key component to a comprehensive knowledge of ocean dynamics and aquatic ecosystems in relation to climate change. However, dynamic features such as fronts are difficult to study through conventional oceanographic techniques. In this paper, we build upon our previous work in sampling and tracking an ocean front based on predictions and/or priors. Specifically, given a prior (that may not be accurate or up-to-date) we present and experimentally validate a method for an autonomous surface or underwater vehicle to plan a mission and adapt this mission on-the-go to track a dynamic, but coherent, structure. Experimental results using a novel indoor testbed, capable of creating controllable fluidic features in an indoor laboratory setting, are presented. These results demonstrate that the vehicle is able to adapt its path to follow a desired, time-varying contour.

A casting manipulator is a robot that launches a payload as a projectile then alters the ballistic trajectory by pulling on an attached tether. Prior work in casting manipulation has focused on applying impulses through the tether to place the payload at a specific location and then retract it. In the spirit of whole-body and whole-arm manipulation, we propose that the tether itself can be used to act on the environment, rather than serving only as a means of directing the projectile. As a first step in this direction, we consider the case of throwing an end-weighted line to wrap a target above and away from the launch point. We explore the necessary conditions for successfully wrapping a target, including the various types of wrapping and the conditions required to make these wraps.

This paper is about life-long vast-scale localisation in spite of changes in weather, lighting and scene structure. Building upon our previous work in Experience-based Navigation [1], we continually grow and curate a visual map of the world that explicitly supports multiple representations of the same place. We refer to these representations as experiences, where a single experience captures the appearance of an environment under certain conditions. Pedagogically, an experience can be thought of as a visual memory. By accumulating experiences we are able to handle cyclic appearance change (diurnal lighting, seasonal changes, and extreme weather conditions) and also adapt to slow structural change. This strategy, although elegant and effective, poses a new challenge: In a region with many stored representations - which one(s) should we try to localise against given finite computational resources? By learning from our previous use of the experience-map, we can make predictions about which memories we should consider next, conditioned on how the robot is currently localised in the experience-map. During localisation, we prioritise the loading of past experiences in order to minimise the expected computation required. We do this in a probabilistic way and show that this memory policy significantly improves localisation efficiency, enabling long-term autonomy on robots with limited computational resources. We demonstrate and evaluate our system over three challenging datasets, totalling 206km of outdoor travel. We demonstrate the system in a diverse range of lighting and weather conditions, scene clutter, camera occlusions, and permanent structural change in the environment.

Human-swarm interaction (HSI) is a developing field of research in which the problem of gesture-based control has been attracting an increasing attention, being at the same time a natural form of interaction and an effective way to point and select individual or groups of robots in the swarm. Gesture-based interaction usually requires vision-based recognition and classification of the gesture from the swarm. At this aim, existing methods for cooperative sensing and recognition make use of distributed consensus algorithms, which include for instance averaging and frequency counting. In this work we present a distributed consensus protocol that allows robot swarms to learn efficiently gestures from online interactions with a human teacher. The protocol also facilitates the integration of different consensus algorithms. Experiments have been performed in emulation using on real data acquired by a swarm of robots. The results indicate that effectively exploiting the collective decision-making of the swarm is a viable way to rapidly achieve good learning performance.

Magnetically actuated micro/nano robots are promising to various biological and medical applications in wet environments. In order to improve robots control, robustness and efficiency, real time 3D topographical surface analysis are essential especially for their navigation under such wet and confined environments. In this paper we propose a swimmer from the Roll-To-Swim (RTS) family capable of detecting and characterizing its surrounding topography thanks to 2D optical feedback. This RTS detects the presence of an obstacle (6.5µm height) by detecting perturbations in its horizontal rolling motion. Then we reconstruct a 2D probabilistic map of obstacles presence. To further characterize the surface topography, we developed the spintop motion where the RTS stands vertically while rotating. We provide a speed characterization of this new motion and demonstrate its maneuverability by integrating it in a closed-loop path following control. In this motion, RTS can be trapped by a targeted obstacle. The amount of power needed to escape is used to characterize the height of the obstacle from 2.5 to 6.5µm with a precision of 2µm. This wireless topographical detection by RTS can be a promising tool to choose an efficient pathway and an appropriate motion adapted to the surrounding environment as well as a surface topography analyzer.

Most wireless solutions today are centered around people-centric devices like laptops and cell phones that are insufficient for mobile robots. The key difference is that people-centric devices use wireless connectivity in bursts under primarily stationary settings while mobile robots continuously transmit data even while moving. When mobile robots use existing wireless solutions, it results in intolerable and seemingly random interruptions in wireless connectivity when moving [1]. These wireless issues stem from suboptimal switching across wireless infrastructure access points (APs), also called AP handoffs. These poor handoff decisions are due to stateless handoff algorithms that make wireless decisions solely from immediate and noisy scans of surrounding wireless conditions. In this paper, we propose to overcome these motion-based wireless connectivity issues for autonomous robots using highly informed handoff algorithms that combine fine-grain wireless maps with accurate robot localization. Our results show significant wireless performance improvements for continuously moving robots in real environments without any modifications to the wireless infrastructure.

Oil and gas refineries can be a dangerous environment for numerous reasons, including heat, toxic gasses, and unexpected catastrophic failures. In order to augment how human operators interact with this environment, a mobile robotic platform is developed. This paper focuses on the use of WiFi for communicating with and localizing the robot. More specifically, algorithms are developed and tested to minimize the total number of WiFi access points (APs) and their locations in any given environment while taking into consideration the throughput requirements and the need to ensure every location in the region can reach at least k APs. When multiple WiFi APs are close together, there is a potential for interference. A graph-coloring heuristic is used to determine AP channel allocation. In addition, WiFi fingerprinting based localization is developed. All the algorithms implemented are tested in real world scenarios with the robot developed and results are promising.

In this paper, we present a novel approach to accurately calibrate the kinematic model of a humanoid based on observations of its monocular camera. Our technique estimates the parameters of the complete model, consisting of the joint angle offsets of the whole body including the legs, as well as the camera extrinsic and intrinsic parameters. We cast the parameter estimation as a least-squares optimization problem. In the error function, we consider the residuals between camera observations of end-effector markers and their projections into the image based on the estimate of the calibration parameters. Furthermore, we developed an approach to automatically select a subset of configurations for the calibration process that yields a good trade-off between the number of observations and accuracy. As the experiments with a Nao humanoid show, we achieve an accurate calibration for this low-cost platform. Further, our approach to configuration selection yields substantially better optimization results compared to randomly chosen viable configurations. Hence, our system only requires a reduced number of configurations to achieve accurate results. Our optimization is general and the implementation, which is available online, can easily be applied to different humanoids.

Humanoid robot is able to execute various behavior to manipulate objects because of high degree-of-freedom around the whole-body. Although hands contact with objects and exert force in ordinary pushing motion by robot, pushing motion contacting with the object at various regions of whole-body has potential for extending the scope of feasible manipulation. We derive the fundamental formulas of humanoid robot in the situation that the external force is applied to the arbitrary region of whole-body, and then propose the method to generate and execute the pushing motion based on the formulas. The proposed method is generalized for enabling to select a contact point with an object from whole-body regions and control the pushing force applied to the sensorless region. In order to verify the effectiveness, we show the experimental result that a lifesized humanoid carries large and heavy objects by pushing with various regions of whole-body.

We propose a novel whole-body multi-modal semiautonomous teleoperation framework for mobile manipulator systems, which consists of: 1) Motion capture and whole-body motion mapping to allow the operator to intuitively teleoperate the mobile manipulator without being constrained by the master interface while also fully exploiting whole-body dexterity; 2) Slave robot autonomous control to allow the mobile manipulator to optimally track the operator's whole-body command, while taking into account the user-slave kinematic dissimilarity (slave robot's joint limit, joint velocity limit, and singularity); and 3) Visuo-haptic-vestibular feedback with HMD (Head Mounted Display) for 3D visual information, wearable cutaneous haptic device for manipulation force feedback, and actuated chair for vestibular feedback to reduce HMD-induced motion sickness. Performance of the proposed framework is validated with simulation of a ROV (remotely operated vehicle) manipulator system and some preliminary user studies.

We present a novel approach to mobile robot search for non-stationary objects in partially known environments. We formulate the search as a path planning problem in an environment where the probability of object occurrences at particular locations is a function of time. We propose to explicitly model the dynamics of the object occurrences by their frequency spectra. Using this spectral model, our path planning algorithm can construct plans that reflect the likelihoods of object locations at the time the search is performed. Three datasets collected over several months containing person and object occurrences in residential and office environments were chosen to evaluate the approach. Several types of spatio-temporal models were created for each of these datasets and the efficiency of the search method was assessed by measuring the time it took to locate a particular object. The results indicate that modeling the dynamics of object occurrences reduces the search time by 25% to 65% compared to maps that neglect these dynamics.

Quickly finding a free parking spot that is close to a desired target location can be a difficult task. This holds for human drivers and autonomous cars alike. In this paper, we investigate the problem of predicting the occupancy of parking spaces and exploiting this information during route planning. We propose an MDP-based planner that considers route information as well as the occupancy probabilities of parking spaces to compute the path that minimizes the expected total time for finding an unoccupied parking space and for walking from the parking location to the target destination. We evaluated our system on real world data gathered over several days in a real parking lot. We furthermore compare our approach to three parking strategies and show that our method outperforms the alternative behaviors.

We address the problem of finding the optimal assignment of tasks to a team of robots when the associated costs may vary, which arises when robots deal with uncertain or dynamic situations. We detail how to compute a sensitivity analysis that characterizes how much costs may change before optimality is violated. Using this analysis, robots are able to avoid unnecessary re-assignment computations and reduce global communication. First, given a model of how costs may evolve, we develop an algorithm to partition the robots into independent cliques, each of which maintains global optimality by communicating only amongst themselves. Second, we propose a method for computing the worst-case sub-optimality if robots persist with the initial assignment, performing no further communication/computation. Lastly, we develop an algorithm that assesses whether cost changes affect the optimality through an escalating succession of local checks. Experiments show that the methods reduce the degree of centralization needed by a multi-robot system.

One open problem in the fields of place recognition and mapping is to be able to recognise a revisited place when its appearance and layout have changed between visits. In this paper, we investigate this problem in the context of RGB-D mapping in indoor environments. We propose to segment the scene in juts (neighbourhood of 3D points with normals that stick out from the surroundings) and look at low-level features, like textureness or entropy of the normals. These could differentiate those zones of the scene that change or move along time from those that are likely to remain static. We also present a method which improves the matching between images of the same place taken at different times by pruning details basing on these features. We evaluate on a number of communal areas and also on some scenes captured 6 months apart. Experiments with our approach, show an increase up to 70% in inlier matching ratio at the cost of pruning only less than 20% of correct matches, without the need of performing geometric verification.

This paper describes the design and manufacturing of soft artificial skin with an array of embedded soft strain sensors for detecting various hand gestures by measuring joint motions of five fingers. The proposed skin was made of a hyperelastic elastomer material with embedded microchannels filled with two different liquid conductors, an ionic liquid and a liquid metal. The ionic liquid microchannels were used to detect the mechanical strain changes of the sensing material, and the liquid metal microchannels were used as flexible and stretchable electrical wires for connecting the sensors to an external control circuit. The two heterogeneous liquid conductors were electrically interfaced through flexible conductive threads to prevent the two liquid from being intermixed. The skin device was connected to a computer through a microcontroller instrumentation circuit for reconstructing the 3-D hand motions graphically. The paper also presents preliminary calibration and experimental results.

Here, we present a general framework for combining visual odometry and lidar odometry in a fundamental and first principle method. The method shows improvements in performance over the state of the art, particularly in robustness to aggressive motion and temporary lack of visual features. The proposed on-line method starts with visual odometry to estimate the ego-motion and to register point clouds from a scanning lidar at a high frequency but low fidelity. Then, scan matching based lidar odometry refines the motion estimation and point cloud registration simultaneously.We show results with datasets collected in our own experiments as well as using the KITTI odometry benchmark. Our proposed method is ranked #1 on the benchmark in terms of average translation and rotation errors, with a 0.75% of relative position drift. In addition to comparison of the motion estimation accuracy, we evaluate robustness of the method when the sensor suite moves at a high speed and is subject to significant ambient lighting changes.

One way to deal with occlusions or loss of tracking of the visual features used for visual servoing tasks is to predict the feature behavior in the image plane when the measurements are missing. Different prediction and correction methods have already been proposed in the literature. The purpose of this paper is to compare and experimentally validate some of these methods for eye-in-hand and eye-to-hand configurations. In particular, we show that a correction based both on the image and the camera/target pose provides the best results.