IROS 2015 Papers

Visual Odometry (VO) is a proven technology for planetary exploration rovers, facilitating their localization with a small error over medium-sized trajectories. However, due to VO's incremental mode of operation, its estimation error accumulates over time, resulting in considerable drift for long trajectories. This paper proposes a global localization method that counters VO drift by matching boulders extracted from overhead and ground images and using them periodically to relocalize the rover and refine VO estimates. The performance of the proposed method is evaluated with the aid of overhead imagery of different resolutions. Experimental results demonstrate that a very terse representation, consisting of approximate boulder locations only, suffices for significantly improving the accuracy of VO over long traverses.

Many modern robotics applications require robots to function autonomously in dynamic environments including other decision making agents, such as people or other robots. This calls for fast and scalable interactive motion planning. This requires models that take into consideration the other agent's intended actions in one's own planning. We present a real-time motion planning framework that brings together a few key components including intention inference by reasoning counterfactually about potential motion of the other agents as they work towards different goals. By using a light-weight motion model, we achieve efficient iterative planning for fluid motion when avoiding pedestrians, in parallel with goal inference for longer range movement prediction. This inference framework is coupled with a novel distributed visual tracking method that provides reliable and robust models for the current belief-state of the monitored environment. This combined approach represents a computationally efficient alternative to previously studied policy learning methods that often require significant offline training or calibration and do not yet scale to densely populated environments. We validate this framework with experiments involving multi-robot and human-robot navigation. We further validate the tracker component separately on much larger scale unconstrained pedestrian data sets.

We consider a problem on the synthesis of optimal reactive controllers with an a priori unknown performance criterion while satisfying a given temporal logic specification through the interaction with an uncontrolled environment. We decouple the problem into two sub-problems. First, we extract a (maximally) permissive strategy for the system, which encodes multiple (possibly all) ways in which the system can react to the adversarial environment and satisfy the specifications. Then, we quantify the a priori unknown performance criterion as a (still unknown) reward function, and compute - by using the so-called maximin-Q learning algorithm - an optimal strategy for the system within the operating envelope allowed by the permissive strategy. We establish both correctness (with respect to the temporal logic specifications) and optimality (with respect to the a priori unknown performance criterion) of this two-step technique for a fragment of temporal logic specifications. For specifications beyond this fragment, correctness can still be preserved, but the learned strategy may be sub-optimal. We present an algorithm to the overall problem, and demonstrate its use and computational requirements on a set of robot motion planning examples.

This paper tackles the problem of coordinated vision-based tracking of a ground target by a fleet of multiple UAVs that exchange information over a supporting time-varying network. The objective of this work is to formulate decentralized control algorithms that enable multiple vehicles to follow a target while coordinating their phase separation. A typical scenario involves multiple aerial surveillance UAVs which are required to monitor a moving ground object (target tracking), while maintaining a desired inter-vehicle separation (coordination). To solve the tracking problem, the yaw rate of each vehicle is used as control input, while the ground speeds are adjusted to achieve coordination. It is assumed that the UAVs are equipped with an internal autopilot, which is able to track yaw rate and ground speed commands. The performance of the coordinated vision-based tracking algorithm is evaluated as a function of target's velocity, tracking performance of the onboard autopilot, and the quality of service of the communication network.

This work tackles the problem of pursuit-evasion games between two pursuing and one evading unmanned aerial vehicle. The solution of this problem is derived by introducing a hierarchical decomposition of the game. On a superordinate collaboration level, the pursuers choose their optimal behavioral strategy (i.e. pursue and herd) resulting in a three-player non-cooperative dynamic game which is solved in a subordinate level of the overall game. This structure enables an intelligent behavior change for the pursuers based on game-theoretical solution methods. Depending on the state of the game it has to be evaluated which behavioral strategy yields the best results for the pursuers within the regarded time horizon. It is shown, that the pursuer's outcome can be improved by using a superordinate cooperation between them. Furthermore, this application was implemented on a real quad-rotor system. In experiments the quad-rotor teamed-up with a software-emulated pursuer and it was shown that another software-emulated quad-rotor was caught. This framework provides a solution concept for all types of dynamic games with an arbitrary number of players and teams having a non-cooperative and cooperative nature. The hierarchical decomposition gives more flexibility and allows team members to co-ordinate their behavior and thus to maximize their outcome.

This paper addresses the problem of cooperative object manipulation with the coordination relying solely on implicit communication. We consider a decentralized leader-follower architecture where the leading robot, that has exclusive knowledge of the object's desired trajectory, tries to achieve the desired tracking behavior via an impedance control law. On the other hand, the follower estimates the leader's desired motion via a novel prescribed performance estimation law, that drives the estimation error to an arbitrarily small residual set, and implements a similar impedance control law. Both control schemes adopt feedback linearization as well as load sharing among the robots according to their specific payload capabilities. The feedback relies exclusively on each robot's force/torque, position as well as velocity measurements and apart from a few commonly predetermined constant parameters, no explicit data is exchanged on-line among the robots, thus reducing the required communication bandwidth and increasing robustness. Finally, a comparative simulation study clarifies the proposed

In this paper, an impedance control scheme for cooperative quadrotors with robotic arms is proposed in order to limit both the contact forces, due to the object/environment interaction, and the internal forces, due to the manipulators/ object interaction. To this aim, two impedance filters are used to determine the reference trajectories for manipulator end effectors: the first is aimed at conferring a compliant behavior at the object level (external impedance), while the second filter, is aimed at avoiding large internal loading of the object (internal impedance). Such trajectories are fed to a motion controller including an inverse kinematics algorithm and a PD controller with gravity compensation. The effectiveness of the proposed approach is then verified in simulation.

In the future mobile dual-arm robots are expected to perform many tasks. Kinematically, the configuration of two manipulators that branch from the same common mobile base results in a serial-to-parallel kinematic structure, which makes inverse kinematic computations non-trivial. The motion of the base has to be decided in a trade-off, taking the needs of both arms into account. We propose to use a Virtual Kinematic Chain (VKC) to specify the common motion of the parallel manipulators, instead of using the two manipulators kinematics directly. With this VKC, we formulate a constraint based programming solution for the robot to respond to external disturbances during task execution. The proposed approach is experimentally verified both in a noise-free illustrative simulation and a real human robot co-manipulation task.

In this paper, a realtime pseudo-structure extraction algorithm for 3D indoor point cloud data (PCD) is proposed. This algorithm is called Convex Cut (CC) because of its two main steps: cutting the PCD with arbitrary planes, and extracting convex parts. CC can be used as a preprocessing module for other existing algorithms to extract static parts in dynamic environments or to represent a principal 3D model of a given PCD. Its calculation time is 24 milliseconds for 50k PCD on a consumer PC, and it yields a precision value of 0.90 and a recall value of 0.99 on average in highly dynamic and cluttered environments. Some possible applications are explained such as simultaneous localization and mapping in dynamic environments, efficient dense map representation, robust 3D scan matching with plane features, and natural motion planning.

A novel control scheme based on wave variables for controlling the teleoperation system over the time-varying delayed communication is proposed. Regarding to the symmetric wave teleoperation control layout, the control is achieved by adjusting the wave variables based on the difference between the desired and the fictitious position, which is obtained from integrating the non-distorted wave variables. The resulting wave is then applied to the master/slave robots controlled with local PD controller. By this control scheme, the system is stable and the position difference of the master and slave is proven to converge to zero in the free motion case if the time delay is continuously differentiable. In addition, the slave-environment interaction force is reflected to the operator precisely when the static contact is made during steady time delay. This control scheme can readily be extended to the implementation of the teleoperation system in full task space SO(3) × ℝ3.

This paper investigates the design and implementation of a finger-like robotic structure capable of reproducing human hand gestural movements performed by a multi-fingered, hand-like structure. In this work, we present a pneumatic circuit and a closed-loop controller for a finger-like soft pneumatic actuator. Experimental results demonstrate the performance of the pneumatic and control systems of the soft pneumatic actuator, and its ability to track human movement trajectories with affective content.

In this paper, we propose a physically meaningful energy-related safety indicator for robots sharing their workspace with humans. Based on this indicator, a safety criterion accounting for the breaking capabilities of the robot is included as a quadratic constraint in the control algorithm. This constraint is modulated by the distance between the human operator and the end-effector of the robot. The control algorithm is formulated as an optimization problem and computes the actuation torque of a robotic manipulator given some task to be performed and physical constraints to respect. The overall framework is validated in a physics simulation software on a Kuka LWR4 and different behaviours of the robot towards a considered obstacle in its environment are evaluated and discussed.

Series elastic actuators primarily use linear springs in their drivetrains, which introduces a design tradeoff: soft springs provide higher torque resolution at the cost of system bandwidth, whereas stiff springs provide a fast response but lower torque resolution. Nonlinear springs (NLSs) potentially incorporate the benefits of both soft and stiff springs, but such springs are often large. An NLS design was recently proposed that combines a variable radius cam with a rubber elastic element, enabling a compact spring design. However, the rubber introduces hysteresis, which can lead to poor torque tracking if not accounted for in the controller. To overcome this limitation, we here propose a state observer that captures hysteretic effects exhibited by the rubber to provide an accurate estimate of actuator torque. We perform torque-control experiments with this observer on an actuator testbed and compare the performance of the NLS to both soft and stiff linear metal springs. Experiments show that the NLS exhibits improved output impedance compared to both linear springs, and comparable bandwidth to the stiff linear spring up to 1.5 Hz. However, the hysteresis in the urethane rubber introduces instability in higher-frequency conditions, suggesting that future NLS designs can be improved by use of a different rubber as the elastic element.

Humanoid robots being underactuated, they need to interact with their environment to move. When optimizing trajectories, the contact forces must therefore be taken into account. In this paper, we first highlight an issue encountered when using parametrized functions in the presence of equality constraints which need to be satisfied continuously over a time interval. We then propose a parametrization of contact forces and a formulation of the constraints which allow to write a tractable optimization program whose solution verifies at any instant all the constraints where those forces appear. We finish by exemplifying the approach with preliminary results obtained on a HRP-2 humanoid robot.

This paper reports on a system for dynamic obstacle tracking for autonomous vehicles. In this work, we seek to simultaneously estimate both the trajectory of the obstacle and the obstacle's shape. These two tasks are inherently coupled-given only noisy partial views, one cannot accurately estimate the trajectory of an obstacle if its shape is unknown, nor can one estimate its shape without knowing its trajectory. To address this challenge, we note that simultaneous localization and mapping (SLAM), where a robot must build a map of the environment while localizing itself within the map, presents similar challenges. By treating the obstacle's shape as a “map” in the obstacle's moving reference frame, we can formulate the obstacle tracking and shape estimation similarly to SLAM. Additionally, we use a continuous time estimation framework to incorporate sensor data that is collected at a fast rate (e.g., light detection and ranging (LIDAR)). Using these methods, we are able to obtain smooth trajectories and crisp point clouds for tracked obstacles. We test our proposed tracker on real-world data collected by our autonomous vehicle platform and demonstrate that it produces improved results when compared to a standard centroid-based extended Kalman filter (EKF) tracker.

Cut along the surface of a polyhedron and unfold it to a planar structure without overlapping is known as Unfolding Polyhedra problem which has been extensively studied in the mathematics literature for centuries. However, whether there exists a continuous unfolding motion such that the polyhedron can be continuously transformed to its unfolding has not been well studied. Recently, researchers started to recognize continuous unfolding as a key step in designing and implementation of self-folding robots. In this paper, we model the unfolding of a polyhedron as multi-link tree-structure articulated robot, and address this problem using motion planning techniques. Instead of sampling in continuous domain which traditional motion planners do, we propose to sample only in the discrete domain. Our experimental results show that sampling in discrete domain is efficient and effective for finding feasible unfolding paths.

Online 3D point cloud classification and scene understanding are crucial tasks for Unmanned Ground Vehicles (UGVs) equipped with multiple laser scanners. Due to the poor performance of traditional 2D image representation model for 3D point clouds, a novel Optimal Bearing Angle (OBA) model is therefore proposed to overcome the limitations of texture information losing and image blurring caused by the UGV's on-the-fly navigation. With the result of 3DSLIC based super-pixel segmentation in OBA images, the center of points belonging to each segmented OBA image patch is assigned as CRF graph node, so that a simplified CRF graph structure is constructed for online contextual classification of 3D laser points in urban environments. Moreover, total 29-dimensional features are extracted from both the raw 3D laser points and the corresponding OBA images. A large number of urban scenes selected from both DUT2 dataset and KAIST dataset are used as testing data in our experiments, and the results show the validity and performance of the proposed method.

In this paper, we propose a new approach for recognizing intentions of humans by observing their activities with an RGB-D camera. Activities and goals are modeled as a distributed network of inter-connected nodes in an Activation Spreading Network (ASN). Inspired by a formalism in hierarchical task networks, the structure of the network captures the hierarchical relationship between high-level goals and low-level activities that realize these goals. Our approach can detect intentions before they are realized and it can work in real-time. We also extend the formalism of ASNs to incorporate contextual information into intent recognition. A fully functioning system is developed for experimental evaluation. We implemented a robotic system that uses our intent recognition to naturally interact with the user. Our ASN based intent recognizer is tested against two different scenarios involving everyday activities performed by a subject, and our results show that the proposed approach is able to detect low-level activities and recognize high-level intentions effectively in real-time. Further analysis shows that contextual ASN is able to discriminate between otherwise ambiguous goals.

For generating whole-body behavior involving contacts with environments such as climbing ladder behavior or walking on terrain behavior, “contact-before-motion” approach was used in some previous researches. By separating contact search process and motion search process, whole-body behavior generation was achieved. However, in previous researches, there were few examinations about contact transition strategies while generating behavior. For changing contact states, there are many kinds of contact transition strategies such as walking, sliding, rotating and so on. For example, while generating standing up behavior, it is important to switch contact transition strategies because the lack of degree of freedom makes it difficult to detach limb contacts from ground, and it may be desirable not to detach contacts but to slide them. In this study, we propose a novel whole-body behavior generation algorithm which involves contact transition strategies switching function. Especially, in this paper, we focus on the walk-type and slide-type transition strategies switching. We call walk-type transition as the contact transition process which detaches some contacts, moves them, and attaches them again. Besides, we call slide-type transition as the contact transition process which keeps on attaching contacts, and slides to move them. By using this algorithm, it is possible to generate whole-body behaviors which are difficult or impossible to achieve only with walk-type transition and which are more desirable by comparing multiple transition strategies. Finally, we evaluated this algorithm by generating standing up behavior and sitting on chair behavior.