Showing papers in "IEEE Transactions on Robotics in 2017"

ORB-SLAM2: An Open-Source SLAM System for Monocular, Stereo, and RGB-D Cameras

Abstract: We present ORB-SLAM2, a complete simultaneous localization and mapping (SLAM) system for monocular, stereo and RGB-D cameras, including map reuse, loop closing, and relocalization capabilities. The system works in real time on standard central processing units in a wide variety of environments from small hand-held indoors sequences, to drones flying in industrial environments and cars driving around a city. Our back-end, based on bundle adjustment with monocular and stereo observations, allows for accurate trajectory estimation with metric scale. Our system includes a lightweight localization mode that leverages visual odometry tracks for unmapped regions and matches with map points that allow for zero-drift localization. The evaluation on 29 popular public sequences shows that our method achieves state-of-the-art accuracy, being in most cases the most accurate SLAM solution. We publish the source code, not only for the benefit of the SLAM community, but with the aim of being an out-of-the-box SLAM solution for researchers in other fields.

SVO: Semidirect Visual Odometry for Monocular and Multicamera Systems

Abstract: Direct methods for visual odometry (VO) have gained popularity for their capability to exploit information from all intensity gradients in the image. However, low computational speed as well as missing guarantees for optimality and consistency are limiting factors of direct methods, in which established feature-based methods succeed instead. Based on these considerations, we propose a semidirect VO (SVO) that uses direct methods to track and triangulate pixels that are characterized by high image gradients, but relies on proven feature-based methods for joint optimization of structure and motion. Together with a robust probabilistic depth estimation algorithm, this enables us to efficiently track pixels lying on weak corners and edges in environments with little or high-frequency texture. We further demonstrate that the algorithm can easily be extended to multiple cameras, to track edges, to include motion priors, and to enable the use of very large field of view cameras, such as fisheye and catadioptric ones. Experimental evaluation on benchmark datasets shows that the algorithm is significantly faster than the state of the art while achieving highly competitive accuracy.

On-Manifold Preintegration for Real-Time Visual--Inertial Odometry

Abstract: Current approaches for visual--inertial odometry (VIO) are able to attain highly accurate state estimation via nonlinear optimization. However, real-time optimization quickly becomes infeasible as the trajectory grows over time; this problem is further emphasized by the fact that inertial measurements come at high rate, hence, leading to the fast growth of the number of variables in the optimization. In this paper, we address this issue by preintegrating inertial measurements between selected keyframes into single relative motion constraints. Our first contribution is a preintegration theory that properly addresses the manifold structure of the rotation group. We formally discuss the generative measurement model as well as the nature of the rotation noise and derive the expression for the maximum a posteriori state estimator. Our theoretical development enables the computation of all necessary Jacobians for the optimization and a posteriori bias correction in analytic form. The second contribution is to show that the preintegrated inertial measurement unit model can be seamlessly integrated into a visual--inertial pipeline under the unifying framework of factor graphs. This enables the application of incremental-smoothing algorithms and the use of a structureless model for visual measurements, which avoids optimizing over the 3-D points, further accelerating the computation. We perform an extensive evaluation of our monocular VIO pipeline on real and simulated datasets. The results confirm that our modeling effort leads to an accurate state estimation in real time, outperforming state-of-the-art approaches.

Safety Barrier Certificates for Collisions-Free Multirobot Systems

Abstract: This paper presents safety barrier certificates that ensure scalable and provably collision-free behaviors in multirobot systems by modifying the nominal controllers to formally satisfy safety constraints. This is achieved by minimizing the difference between the actual and the nominal controllers subject to safety constraints. The resulting computation of the safety controllers is done through a quadratic programming problem that can be solved in real-time and in this paper, we describe a series of problems of increasing complexity. Starting with a centralized formulation, where the safety controller is computed across all agents simultaneously, we show how one can achieve a natural decentralization whereby individual robots only have to remain safe relative to nearby robots. Conservativeness and existence of solutions as well as deadlock-avoidance are then addressed using a mixture of relaxed control barrier functions, hybrid braking controllers, and consistent perturbations. The resulting control strategy is verified experimentally on a collection of wheeled mobile robots whose nominal controllers are explicitly designed to make the robots collide.

Robot Collisions: A Survey on Detection, Isolation, and Identification

Abstract: Robot assistants and professional coworkers are becoming a commodity in domestic and industrial settings. In order to enable robots to share their workspace with humans and physically interact with them, fast and reliable handling of possible collisions on the entire robot structure is needed, along with control strategies for safe robot reaction. The primary motivation is the prevention or limitation of possible human injury due to physical contacts. In this survey paper, based on our early work on the subject, we review, extend, compare, and evaluate experimentally model-based algorithms for real-time collision detection, isolation, and identification that use only proprioceptive sensors. This covers the context-independent phases of the collision event pipeline for robots interacting with the environment, as in physical human–robot interaction or manipulation tasks. The problem is addressed for rigid robots first and then extended to the presence of joint/transmission flexibility. The basic physically motivated solution has already been applied to numerous robotic systems worldwide, ranging from manipulators and humanoids to flying robots, and even to commercial products.

Proprioceptive Actuator Design in the MIT Cheetah: Impact Mitigation and High-Bandwidth Physical Interaction for Dynamic Legged Robots

Abstract: Designing an actuator system for highly dynamic legged robots has been one of the grand challenges in robotics research. Conventional actuators for manufacturing applications have difficulty satisfying design requirements for high-speed locomotion, such as the need for high torque density and the ability to manage dynamic physical interactions. To address this challenge, this paper suggests a proprioceptive actuation paradigm that enables highly dynamic performance in legged machines. Proprioceptive actuation uses collocated force control at the joints to effectively control contact interactions at the feet under dynamic conditions. Modal analysis of a reduced leg model and dimensional analysis of DC motors address the main principles for implementation of this paradigm. In the realm of legged machines, this paradigm provides a unique combination of high torque density, high-bandwidth force control, and the ability to mitigate impacts through backdrivability. We introduce a new metric named the “impact mitigation factor” (IMF) to quantify backdrivability at impact, which enables design comparison across a wide class of robots. The MIT Cheetah leg is presented, and is shown to have an IMF that is comparable to other quadrupeds with series springs to handle impact. The design enables the Cheetah to control contact forces during dynamic bounding, with contact times down to 85 ms and peak forces over 450 N. The unique capabilities of the MIT Cheetah, achieving impact-robust force-controlled operation in high-speed three-dimensional running and jumping, suggest wider implementation of this holistic actuation approach.

Interactive Perception: Leveraging Action in Perception and Perception in Action

Abstract: Recent approaches in robot perception follow the insight that perception is facilitated by interaction with the environment. These approaches are subsumed under the term Interactive Perception (IP). This view of perception provides the following benefits. First, interaction with the environment creates a rich sensory signal that would otherwise not be present. Second, knowledge of the regularity in the combined space of sensory data and action parameters facilitates the prediction and interpretation of the sensory signal. In this survey, we postulate this as a principle for robot perception and collect evidence in its support by analyzing and categorizing existing work in this area. We also provide an overview of the most important applications of IP. We close this survey by discussing remaining open questions. With this survey, we hope to help define the field of Interactive Perception and to provide a valuable resource for future research.

Passive Particle Jamming and Its Stiffening of Soft Robotic Grippers

Abstract: The compliance of soft grippers contributes to their great superiority over rigid grippers in grasping irregularly shaped objects and forming soft contact with environments. Due to a relatively small pressure, soft grippers lack the stiffness required for wider applications. Particle jamming has been frequently reported as a means of stiffness control. Unlike previous research using vacuum for particle jamming, this paper proposes a novel passive particle jamming principle that does not need any vacuum power or other control means. The proposed method is by simply patching a silicone rubber soft actuator and a pack (made of strain-limiting membrane) of particles to form an integral gripping finger. The inflation of the soft actuator applies a pressure to the particle pack causing particles inside it to jam. A larger squeezing pressure will result in tighter particle jamming, thus increasing the stiffness of the finger. The stiffness of the finger is controllable as it is proportional to the actuator's air pressure, which has been verified by experiments in this research. The stiffness can increase more than six fold when air pressure changes from 20 to 80 kPa in the experimental studies. The reported discovery may enhance the capabilities of soft robotic grippers so that more robotic picking operations could be performed by soft grippers.

How Behavior Trees Modularize Hybrid Control Systems and Generalize Sequential Behavior Compositions, the Subsumption Architecture, and Decision Trees

Abstract: Behavior trees (BTs) are a way of organizing the switching structure of a hybrid dynamical system (HDS), which was originally introduced in the computer game programming community. In this paper, we analyze how the BT representation increases the modularity of an HDS and how key system properties are preserved over compositions of such systems, in terms of combining two BTs into a larger one. We also show how BTs can be seen as a generalization of sequential behavior compositions, the subsumption architecture, and decisions trees. These three tools are powerful but quite different, and the fact that they are unified in a natural way in BTs might be a reason for their popularity in the gaming community. We conclude the paper by giving a set of examples illustrating how the proposed analysis tools can be applied to robot control BTs.

RGB-D Object Recognition and Grasp Detection Using Hierarchical Cascaded Forests

Abstract: This paper presents an efficient framework to perform recognition and grasp detection of objects from RGB-D images of real scenes. The framework uses a novel architecture of hierarchical cascaded forests, in which object-class and grasp-pose probabilities are computed at different levels of an image hierarchy (e.g., patch and object levels) and fused to infer the class and the grasp of unseen objects. We introduce a novel training objective function that minimizes the uncertainties of the class labels and the grasp ground truths at the leaves of the forests, thereby enabling the framework to perform the recognition and grasp detection of objects. Our objective function is learned from features that are extracted from RGB-D point clouds of the objects. For that, we propose a novel method to encode an RGB-D point cloud into a representation that facilitates the use of large convolution neural networks to extract discriminative features from RGB-D images. We evaluate our framework on challenging object datasets, where we demonstrate that our framework outperforms the state-of-the-art methods in terms of object-recognition and grasp-detection accuracies. We also show experiments by using live video streams from a Kinect mounted on our in-house robotic platform.

Adaptive Human–Robot Interaction Control for Robots Driven by Series Elastic Actuators

Abstract: Series elastic actuators (SEAs) are known to offer a range of advantages over stiff actuators for human-robot interaction, such as high force/torque fidelity, low impedance, and tolerance to shocks. While a variety of SEAs have been developed and implemented in initiatives that involve physical interactions with humans, relatively few control schemes were proposed to deal with the dynamic stability and uncertainties of robotic systems driven by SEAs, and the open issue of safety that resolves the conflicts of motion between the human and the robot has not been systematically addressed. In this paper, a novel continuous adaptive control method is proposed for SEA-driven robots used in human-robot interaction. The proposed method provides a unified formulation for both the robot-in-charge mode, where the robot plays a dominant role to follow a desired trajectory, and the human-in-charge mode, in which the human plays a dominant role to guide the movement of robot. Instead of designing multiple controllers and switching between them, both typical modes are integrated into a single controller, and the transition between two modes is smooth and stable. Therefore, the proposed controller is able to detect the human motion intention and guarantee the safe human-robot interaction. The dynamic stability of the closed-loop system is theoretically proven by using the Lyapunov method, with the consideration of uncertainties in both the robot dynamics and the actuator dynamics. Both simulation and experimental results are presented to illustrate the performance of the proposed controller.

FreMEn: Frequency Map Enhancement for Long-Term Mobile Robot Autonomy in Changing Environments

Abstract: We present a new approach to long-term mobile robot mapping in dynamic indoor environments. Unlike traditional world models that are tailored to represent static scenes, our approach explicitly models environmental dynamics. We assume that some of the hidden processes that influence the dynamic environment states are periodic and model the uncertainty of the estimated state variables by their frequency spectra. The spectral model can represent arbitrary timescales of environment dynamics with low memory requirements. Transformation of the spectral model to the time domain allows for the prediction of the future environment states, which improves the robot's long-term performance in changing environments. Experiments performed over time periods of months to years demonstrate that the approach can efficiently represent large numbers of observations and reliably predict future environment states. The experiments indicate that the model's predictive capabilities improve mobile robot localization and navigation in changing environments.

On the Control and Properties of Supercoiled Polymer Artificial Muscles

Abstract: Robotics have long sought an actuation technology comparable to or as capable as biological muscle tissue. Natural muscles exhibit a high power-to-weight ratio, inherent compliance and damping, fast action, and a high dynamic range. They also produce joint displacements and forces without the need for gearing or additional hardware. Recently, supercoiled commercially available polymer threads (sewing thread or nylon fishing lines) have been used to create significant mechanical power in a muscle-like form factor. Heating and cooling the polymer threads causes contraction and expansion, which can be utilized for actuation. In this paper, we describe the working principle of supercoiled polymer (SCP) actuation and explore the controllability and properties of these threads. We show that under appropriate environmental conditions, the threads are suitable as a building block for a controllable artificial muscle. We leverage off-the-shelf silver-coated threads to enable rapid electrical heating while the low thermal mass allows for rapid cooling. We utilize both thermal and thermomechanical models for feed-forward and feedback control. The resulting SCP actuator regulates to desired force levels in as little as 28 ms. Together with its inherent stiffness and damping, this is sufficient for a position controller to execute large step movements in under 100 ms. This controllability, high performance, the mechanical properties, and the extremely low material cost are indicative of a viable artificial muscle.

Modeling of Viscoelastic Electromechanical Behavior in a Soft Dielectric Elastomer Actuator

Abstract: Soft dielectric elastomer actuators (DEAs) exhibit interesting muscle-like behavior for the development of soft robots. However, it is challenging to model these soft actuators due to their material nonlinearity, nonlinear electromechanical coupling, and time-dependent viscoelastic behavior. Most recent studies on DEAs focus on issues of mechanics, physics, and material science, while much less importance is given to quantitative characterization of DEAs. In this paper, we present a detailed experimental investigation probing the voltage-induced electromechanical response of a soft DEA that is subjected to cyclic loading and propose a general constitutive modeling approach to characterize the time-dependent response, based on the principles of nonequilibrium thermodynamics. In this paper, some of the key observations are found as follows: 1) Creep exhibits the drift phenomenon, and is dominant during the first three cycles. The creep decreases over time and becomes less dominant after the first few cycles; 2) a significant amount of hysteresis is observed during all cycles and it becomes repeatable after the first few cycles; 3) the peak of the displacement is shifted from the peak of the voltage signal and occurs after it. To account for these viscoelastic phenomena, a constitutive model is developed by employing several dissipative nonequilibrium mechanisms. The quantitative comparisons of the experimental and simulation results demonstrate the effectiveness of the developed model. This modeling approach can be useful for control of a viscoelastic DEA and paves the way to emerging applications of soft robots.

A Multimodal Robot for Perching and Climbing on Vertical Outdoor Surfaces

Abstract: Perching can extend the useful mission life of a micro air vehicle. Once perched, climbing allows it to reposition precisely, with low power draw and without regard for weather conditions. We present the Stanford Climbing and Aerial Maneuvering Platform, which is to our knowledge the first robot capable of flying, perching with passive technology on outdoor surfaces, climbing, and taking off again. We present the mechanical design and the new perching, climbing, and takeoff strategies that allow us to perform these tasks on surfaces such as concrete and stucco, without the aid of a motion capture system or off-board computation. We further discuss two new capabilities uniquely available to a hybrid aerial–scansorial robot: the ability to recover gracefully from climbing failures and the ability to increase usable foothold density through the application of aerodynamic forces. We also measure real power consumption for climbing, flying, and monitoring and discuss how future platforms could be improved for longer mission life.

A Distributed Version of the Hungarian Method for Multirobot Assignment

Abstract: In this paper, we propose a distributed version of the Hungarian method to solve the well-known assignment problem. In the context of multirobot applications, all robots cooperatively compute a common assignment that optimizes a given global criterion (e.g., the total distance traveled) within a finite set of local computations and communications over a peer-to-peer network. As a motivating application, we consider a class of multirobot routing problems with “spatiotemporal” constraints, i.e., spatial targets that require servicing at particular time instants. As a means of demonstrating the theory developed in this paper, the robots cooperatively find online suboptimal routes by applying an iterative version of the proposed algorithm in a distributed and dynamic setting. As a concrete experimental test bed, we provide an interactive “multirobot orchestral” framework, in which a team of robots cooperatively plays a piece of music on a so-called orchestral floor.

Hybrid Dynamic Moving Obstacle Avoidance Using a Stochastic Reachable Set-Based Potential Field

Abstract: One of the primary challenges for autonomous robotics in uncertain and dynamic environments is planning and executing a collision-free path. Hybrid dynamic obstacles present an even greater challenge as the obstacles can change dynamics without warning and potentially invalidate paths. Artificial potential field (APF)-based techniques have shown great promise in successful path planning in highly dynamic environments due to their low cost at runtime. We utilize the APF framework for runtime planning but leverage a formal validation method, Stochastic Reachable (SR) sets, to generate accurate potential fields for moving obstacles. A small number of SR sets are computed a priori , then used to generate a potential field that represents the obstacle's stochastic motion for online path planning. Our method is novel and scales well with the number of obstacles, maintaining a relatively high probability of reaching the goal without collision, as compared to other traditional Gaussian APF methods. Here, we demonstrate our method with up to 900 hybrid dynamic obstacles and show that it outperforms the traditional Gaussian APF method by up to 60% in the holonomic case and up to 20% in the unicycle case.

ZMP Support Areas for Multicontact Mobility Under Frictional Constraints

Abstract: We propose a method for checking and enforcing multicontact stability based on the zero-tilting moment point (ZMP). The key to our development is the generalization of ZMP support areas to take into account: 1) frictional constraints and 2) multiple noncoplanar contacts. We introduce and investigate two kinds of ZMP support areas. First, we characterize and provide a fast geometric construction for the support area generated by valid contact forces, with no other constraint on the robot motion. We call this set the full support area. Next, we consider the control of humanoid robots by using the linear pendulum mode (LPM). We observe that the constraints stemming from the LPM induce a shrinking of the support area, even for walking on horizontal floors. We propose an algorithm to compute the new area, which we call the pendular support area. We show that, in the LPM, having the ZMP in the pendular support area is a necessary and sufficient condition for contact stability. Based on these developments, we implement a whole-body controller and generate feasible multicontact motions where an HRP-4 humanoid locomotes in challenging multicontact scenarios.

Anthropomorphic Low-Inertia High-Stiffness Manipulator for High-Speed Safe Interaction

Abstract: In this paper, a manipulator is proposed for safe human–robot interaction at high speed. The manipulator has both low mass and inertia and high stiffness and strength. It is basically a cable-driven manipulator; nevertheless, by using a unique lightweight tension-amplification mechanism, the manipulator retains high stiffness and strength. The joint stiffness, which is strongly related to the motion control performance, is amplified by the quadratic order. Both 1-degree of freedom (DOF) and 3-DOF joint mechanisms using the tension-amplification mechanism are presented and combined to develop a 7-DOF anthropomorphic manipulator named LIMS. The mass and inertia beyond the shoulder were 2.24 kg and 0.599 kg·m2, respectively, which are lower than those of a human. The stiffness of the developed elbow joint was 1410 N·m/rad, which is approximately seven times higher than that of a human. Considering the ratio of stiffness to inertia, the manipulator is expected to show a control performance that is comparable to those of conventional industrial manipulators. Comprehensive experiments, including joint stiffness tests and high-speed interaction tests, were conducted to verify the feasibility of the developed manipulator.

Integral Line-of-Sight Guidance for Path Following Control of Underwater Snake Robots: Theory and Experiments

Abstract: This paper proposes and experimentally validates a straight line path following controller for underwater snake robots in the presence of constant irrotational currents of unknown direction and magnitude. An integral line-of-sight guidance law is presented, which is combined with a sinusoidal gait pattern and a directional controller that steers the robot toward and along the desired path. The stability of the proposed control scheme in the presence of ocean currents is investigated by using Poincare map analysis. Simulation results are presented to illustrate the performance of the proposed path following controller for both lateral undulation and eel-like motion. In addition, the performance of the path following controller is investigated through experiments with a physical underwater snake robot. The experimental results show that the proposed control strategy successfully steers the robot toward and along the desired path in the presence of an unknown constant irrotational current in the inertial frame.

Dynamic In-Hand Sliding Manipulation

Abstract: This paper presents a framework for planning the motion of an $n$ -fingered robot hand to create an inertial load on a grasped object to achieve a desired in-grasp sliding motion. The model of the sliding dynamics is based on a soft-finger limit surface contact model at each fingertip. A motion planner is derived to automatically solve for the finger motions for a given initial and desired configuration of the object relative to the fingers. Iterative planning and execution are shown to reduce the errors that occur due to the modeling and trajectory tracking errors. The framework is applied to the problem of regrasping a laminar object held in a pinch grasp. We propose a limited surface model of the contact pressure distribution at each finger to predict the sliding directions. Experimental validations are shown, including iterative error reduction and repeatability of the experiment.

Probabilistic and Distributed Control of a Large-Scale Swarm of Autonomous Agents

Abstract: We present a distributed control algorithm simultaneously solving both the stochastic target assignment and optimal motion control for large-scale swarms to achieve complex formation shapes. Our probabilistic swarm guidance using inhomogeneous Markov chains (PSG–IMC) algorithm adopts a Eulerian density-control framework, under which the physical space is partitioned into multiple bins and the swarm's density distribution over each bin is controlled in a probabilistic fashion to efficiently handle loss or the addition of agents. We assume that the number of agents is much larger than the number of bins and that each agent knows in which bin it is located, the desired formation shape, and the objective function and motion constraints. PSG–IMC determines the bin-to-bin transition probabilities of each agent using a time IMC. These time-varying Markov matrices are computed by each agent in real time using the feedback from the current swarm distribution, which is estimated in a distributed manner. The PSG–IMC algorithm minimizes the expected cost of transitions per time instant that are required to achieve and maintain the desired formation shape, even if agents are added to or removed from the swarm. PSG–IMC scales well with a large number of agents and complex formation shapes and can also be adapted for area exploration applications. We demonstrate the effectiveness of this proposed swarm guidance algorithm by using numerical simulations and hardware experiments with multiple quadrotors.

Magnetic Actuation for Full Dexterity Microrobotic Control Using Rotating Permanent Magnets

Abstract: Recent work in magnetically actuated microscale robots for biomedical or microfluidic applications has resulted in magnetic actuation systems that can remotely command precise five-degree-of-freedom control of magnetic devices. This paper presents a new type of actuation system, which uses an array of rotating permanent magnets to generate the same level of control over untethered microscale devices with the potential for increased field and gradient strength and minimal heat generation. In contrast with previous permanent-magnet actuation systems, the system proposed here does not require any hazardous translational motion of the control magnets, resulting in a simple, safe, and inexpensive system. The proof-of-concept prototype system presented, with eight permanent magnets, can create fields and field gradients in any direction with variable magnitudes between zero and 30 mT and ${0.83\,\text{Tm}}^{- 1}$ , respectively. The effectiveness of the system is shown through characterization and feedback control of a 250-μm micromagnet in a 3-D path-following task with average accuracy of 25 μm. An optimization framework is presented for designing system configurations for targeted applications.

On-Demand Disassembly of Paramagnetic Nanoparticle Chains for Microrobotic Cargo Delivery

Abstract: Paramagnetic nanoparticles are considered as attractive building blocks, particularly for robotic delivery of drugs. Although paramagnetic nanoparticles can be effectively gathered and transported using external magnetic fields, the disassembly process is yet to be fully investigated to avoid the formation of aggregations. In this paper, we report a novel method of controllable disassembly of paramagnetic nanoparticle chains using a predefined dynamic magnetic field. The dynamic field is capable of performing spreading and fragmentation of the particle chains simultaneously. Using the magnetic dipole–dipole repulsive forces, the final area covered by the particle chains swells up to 545% of the initial area. The final length distribution presents a strong relationship with the frequency of the dynamic field in deionized (DI) water and two kinds of biofluids. An analytical model of phase lag is proposed, which shows good agreement with the experimental results. Furthermore, we also present an assembly process using a rotating magnetic field, indicating that the assembly disassembly process is reversible. In addition, batch-cargo delivery of polystyrene microbeads using the nanoparticle chains as swarm-like nanorobots is demonstrated.

Stiffness Control With Shape Memory Polymer in Underactuated Robotic Origamis

Abstract: Underactuated systems offer compact design with easy actuation and control but at the cost of limited stable configurations and reduced dexterity compared to the directly driven and fully actuated systems. Here, we propose a compact origami-based design in which we can modulate the material stiffness of the joints and thereby control the stable configurations and the overall stiffness in an underactuated robot. The robotic origami, robogami, design uses multiple functional layers in nominally two-dimensional robots to achieve the desired functionality. To control the stiffness of the structure, we adjust the elastic modulus of a shape memory polymer using an embedded customized stretchable heater. We study the actuation of a robogami finger with three joints and determine its stable configurations and contact forces at different stiffness settings. We monitor the configuration of the finger using feedback from customized curvature sensors embedded in each joint. A scaled down version of the design is used in a two-fingered gripper and different grasp modes are achieved by activating different sets of joints.

Dexterous Aerial Robots—Mobile Manipulation Using Unmanned Aerial Systems

Abstract: In this paper, we present selected benchmark aerial manipulation tasks using an aerial vehicle endowed with multi-degree of freedom manipulators. The proposed tasks analyze environmental coupling and are broken into three general categories: momentary, loose, and strong coupling. A classical control structure is derived, tuned, and verified through experiments, conducted for benchmarking purposes to include pick-and-place, insertion, and valve-turning tasks. Although other nonlinear controllers may prove more effective, the classical control approach has been selected in order to analyze contact stability and provide benchmark results for future reference. An analysis of system stability is conducted and implemented into the controller. A vision-based high-level controller fuses motion tracking data in order to provide control of both the aircraft and the manipulators, allowing the system to become coupled to the environment and perform the required operation. We present recent results validating our framework using the proposed aircraft-arm system.

External Wrench Estimation, Collision Detection, and Reflex Reaction for Flying Robots

Abstract: Flying in unknown environments may lead to unforeseen collisions, which may cause serious damage to the robot and/or its environment. In this context, fast and robust collision detection combined with safe reaction is, therefore, essential and may be achieved using external wrench information. Also, deliberate physical interaction requires a control loop designed for such a purpose and may require knowledge of the contact wrench. In principle, the external wrench may be measured or estimated. Whereas measurement poses large demands on sensor equipment, additional weight, and overall system robustness, in this paper we present a novel model-based method for external wrench estimation in flying robots. The algorithm is based on the onboard inertial measurement unit and the robot's dynamics model only. We design admittance and impedance controllers that use this estimate for sensitive and robust physical interaction. Furthermore, the performance of several collision detection and reaction schemes is investigated in order to ensure collision safety. The identified collision location and associated normal vector located on the robot's convex hull may then be used for sensorless tactile sensing. Finally, a low-level collision reflex layer is provided for flying robots when obstacle avoidance fails, also under wind influence. Our experimental and simulation results show evidence that the methodologies are easily implemented and effective in practice.

Bounded Control Law for Global Connectivity Maintenance in Cooperative Multirobot Systems

Abstract: In this paper, we address the connectivity maintenance problem for a multirobot system that moves according to a given bounded collective control objective. We assume that the interaction among the robotic units is limited by a given visibility radius both in terms of sensing and communication capabilities. For this scenario, we propose a decentralized bounded control law that can provably preserve the connectivity of the multirobot system over time. We characterize the effect of the connectivity control term on the achievement of the collective control objective by resorting to an input-to-state stability-like analysis. We provide numerical and experimental results to corroborate the theoretical findings and assess the effectiveness of the proposed bounded connectivity maintenance control law.

The Impact of Diversity on Optimal Control Policies for Heterogeneous Robot Swarms

Abstract: We consider the problem of distributing a large group of heterogeneous robots among a set of tasks that require specialized capabilities in order to be completed. We model the system of heterogeneous robots as a community of species, in which each species (robot type) is defined by the traits (capabilities) that it owns. In order to solve the distribution problem, we develop centralized as well as decentralized methods to efficiently control the heterogeneous swarm of robots. Our methods assume knowledge of the underlying task topology and are based on a continuous model of the system that defines transition rates to and from tasks, for each robot species. Our optimization of the transition rates is fully scalable with respect to the number of robots, number of species, and number of traits. Building on this result, we propose a real-time optimization method that enables an online adaptation of transition rates as a function of the state of the current robot distribution. We also show how the robot distribution can be approximated based on local information only, consequently enabling the development of a decentralized controller. We evaluate our methods by means of microscopic simulations and show how the performance of the latter is well predicted by the macroscopic equations. Importantly, our framework also includes a diversity metric that enables an evaluation of the impact of swarm heterogeneity on performance. The metric defines the notion of minspecies , i.e., the minimum set of species that are required to achieve a given goal. We show that two distinct goal functions lead to two specializations of minspecies, which we term as eigenspecies and coverspecies . Quantitative results show the relation between diversity and performance.

Elastic Stability of Cosserat Rods and Parallel Continuum Robots

Abstract: Classic theories in nonlinear elasticity have increasingly been used to obtain accurate and efficient models for continuum robots and other elastic structures. Numerically computed solutions of these models typically satisfy the first-order conditions necessary for equilibrium, but do not provide any information about the elastic stability of the solution. The inability to detect or avoid physically unstable model solutions poses a major hindrance to reliable model-based simulation, planning, design, and control. In this paper, we adapt results from optimal control to determine the stability of Kirchhoff rods and Cosserat rods subject to general end constraints, including coupled multirod models which describe parallel continuum robots. We formulate a sufficient condition for the stability of a solution, a numerical test for evaluating this condition, and a heuristic stability metric. We verify that our numerical stability test agrees with the classical results for the buckling of single columns with various end constraints and for multicolumn frames. We then validate our approach experimentally on a six degree-of-freedom parallel continuum robot.