ICRA 2018 论文列表

Electrical motor and hydraulic actuation widely-used in robotics are “internal actuation” with their actuators sitting at the joint between two links. This internal actuation is fundamentally limiting to construct a large-size dexterously-articulated robot, since any external force (and its own link weight) is to be accumulated to the base multiplied by the moment arm length, requiring extremely strong/sturdy base actuator/structure as the system size increases. In this paper, we propose a novel robotic system, LASDRA (large-size aerial skeleton with distributed rotor actuation), which, by utilizing distributed rotors as “external actuation”, can overcome this limitation of internal actuation and enables us to realize large-size dexterously-articulated robots. We present its design and modeling, joint locking strategy to increase its loading capability, and also a novel decentralized control scheme to allow for compliant operation with scalability against the number of links. Trajectory tracking and valve turning experiments are also performed to validate the theory.

In this paper we present KnowRob2, a second generation knowledge representation and reasoning framework for robotic agents. KnowRob2 is an extension and partial redesign of KnowRob, currently one of the most advanced knowledge processing systems for robots that has enabled them to successfully perform complex manipulation tasks such as making pizza, conducting chemical experiments, and setting tables. The knowledge base appears to be a conventional first-order time interval logic knowledge base, but it exists to a large part only virtually: many logical expressions are constructed on demand from data structures of the control program, computed through robotics algorithms including ones for motion planning and solving inverse kinematics problems, and log data stored in noSQL databases. Novel features and extensions of KnowRob2 substantially increase the capabilities of robotic agents of acquiring open-ended manipulation skills and competence, reasoning about how to perform manipulation actions more realistically, and acquiring commonsense knowledge.

This paper discusses the kinematic design and geometric optimization of a novel hybrid three degree-of-freedom (DOF) wrist mechanism. The architecture consists of a one prismatic-revolute-universal linkage and one prismatic-spherical-spherical linkage in parallel with a revolute-universal linkage. This architecture is capable of spherical motion identical to that of a pitch-yaw-roll wrist. Moreover, this mechanism is considered to be partially decoupled, as not all actuators contribute to motion in an arbitrary direction. The forward and inverse kinematics of the parallel 2-DOF mechanism are presented. The 2-DOF mechanism is geometrically optimized over its design parameters to maximize a global transmission index, which measures the motion and torque transmissibility of particular wrist configuration over its workspace. The decoupled nature of the mechanism allows the pitch and yaw mechanism to be optimized separately, greatly reducing the parameter search space and allowing a much larger number of mechanism configurations to be simulated. We leverage this increase in simulated configurations to examine the effect of size constraints on the resulting mechanisms as well.

We introduce a method to optimize on a patient-specific basis the kinematic design of the Continuum Reconfigurable Incisionless Surgical Parallel (CRISP) robot, a needle-diameter medical robot based on a parallel structure that is capable of performing minimally invasive procedures. Our objective is to maximize the ability of the robot's tip camera to view tissue surfaces in constrained spaces. The kinematic design of the CRISP robot, which greatly influences its ability to perform a task, includes parameters that are fixed before the procedure begins, such as entry points into the body and parallel structure connection points. We combine a global stochastic optimization algorithm, Adaptive Simulated Annealing (ASA), with a motion planner designed specifically for the CRISP robot. ASA facilitates exploration of the robot's design space while the motion planner enables evaluation of candidate designs based on their ability to successfully view target regions on a tissue surface. By leveraging motion planning, we ensure that the evaluation of a design only considers motions which do not collide with the patient's anatomy. We analytically show that the method asymptotically converges to a globally optimal solution and demonstrate our algorithm's ability to optimize kinematic designs of the CRISP robot on a patient-specific basis.

We present “just-in-time reconstruction” as realtime image-guided inpainting of a map with arbitrary scale and sparsity to generate a fully dense depth map for the image. In particular, our goal is to inpaint a sparse map - obtained from either a monocular visual SLAM system or a sparse sensor - using a single-view depth prediction network as a virtual depth sensor. We adopt a fairly standard approach to data fusion, to produce a fused depth map by performing inference over a novel fully-connected Conditional Random Field (CRF) which is parameterized by the input depth maps and their pixel-wise confidence weights. Crucially, we obtain the confidence weights that parameterize the CRF model in a data-dependent manner via Convolutional Neural Networks (CNNs) which are trained to model the conditional depth error distributions given each source of input depth map and the associated RGB image. Our CRF model penalises absolute depth error in its nodes and pairwise scale-invariant depth error in its edges, and the confidence-based fusion minimizes the impact of outlier input depth values on the fused result. We demonstrate the flexibility of our method by real-time inpainting of ORB-SLAM, Kinect, and LIDAR depth maps acquired both indoors and outdoors at arbitrary scale and varied amount of irregular sparsity.

When driving in urban environments, an autonomous vehicle must account for the interaction with other traffic participants. It must reason about their future behavior, how its actions affect their future behavior, and potentially consider multiple motion hypothesis. In this paper we introduce a method for joint behavior estimation and trajectory planning that models interaction and multi-policy decision-making. The method leverages Partially Observable Markov Decision Processes to estimate the behavior of other traffic participants given the planned trajectory for the ego-vehicle, and Receding-Horizon Control for generating safe trajectories for the ego-vehicle. To achieve safe navigation we introduce chance constraints over multiple motion policies in the receding-horizon planner. These constraints account for uncertainty over the behavior of other traffic participants. The method is capable of running in real-time and we show its performance and good scalability in simulated multi-vehicle intersection scenarios.

The ability to perceive and predict future positions of dynamic objects is essential for mobile robots and intelligent vehicles in dynamic environments. In this paper, we present a novel planning-based approach for long-term human motion prediction that accounts for local interactions and can accurately predict joint motion of multiple agents. Long-term predictions are handled using an MDP formulation that computes a set of stochastic motion policies. To obtain distributions over future motion trajectories, we sample the policies with a weighted random walk algorithm in which each person is locally influenced by social forces from other nearby agents. Unlike related work, the algorithm is environment-aware, can account for individual agent velocities, requires no training phase and makes joint predictions for multiple agents. Experiments in simulation and with real data show that our method makes more accurate predictions than two state-of-the-art methods in terms of probabilistic and geometrical performance measures.

For a safe and successful daily living assistance, far from the highly controlled environment of a factory, robots should be able to adapt to ever-changing situations. Programming such a robot is a tedious process that requires expert knowledge. An alternative is to rely on a high-level planner, but the generic symbolic representations used are not well suited to particular robot executions. Contrarily, motion primitives encode robot motions in a way that can be easily adapted to different situations. This paper presents a combined framework that exploits the advantages of both approaches. The number of required symbolic states is reduced, as motion primitives provide “smart actions” that take the current state and cope online with variations. Symbolic actions can include interactions (e.g., ask and inform) that are difficult to demonstrate. We show that the proposed framework can adapt to the user preferences (in terms of robot speed and robot verbosity), can readjust the trajectories based on the user movements, and can handle unforeseen situations. Experiments are performed in a shoe-dressing scenario. This scenario is particularly interesting because it involves a sufficient number of actions, and the human-robot interaction requires the handling of user preferences and unexpected reactions.

In-hand manipulation has attracted attention because of its potential for performing dexterous manipulation tasks. Few successful examples using real robotic fingers have been reported because model-based approaches have been assumed. A gradient descent-based iterative learning control is one of the typical methods for improving the control performance without the need for a precise model. However, the learning performances deteriorate greatly owing to the stochastic uncertainties, and the learning rates have to be determined manually. We propose a novel iterative learning scheme with adaptive learning rate methods for dexterous in-hand manipulation. The proposed scheme not only eliminates the need for a precise model and manual tuning of a learning rate but also is robust to stochastic uncertainties and insensitive to hyperparameters. The validity of the proposed iterative learning scheme is demonstrated through several experiments.

Rapid acceleration manoeuvres have been avoided by researchers due to the aperiodicity and complexities of this motion. With the recent improvements in optimal control, this paper presents the first examination of a biped completing a time optimal sprint, starting and ending in rest, to provide insight for parameter choices of a robotic platform. In particular, a realistic linkage morphology is used with the limitation of a pre-specified actuator to choose the nominal leg length and gear ratio. Due to the size of the optimisation problem, a brute force approach is used rather than including these parameters as free variables. The results provided unique motion trajectories for time optimal behaviour with the models reaching near steady state motion and performing manoeuvres that are seen in a biped's biological counterpart. We then show that access to a higher mass-specific force does not improve the rapid acceleration manoeuvres, rather the friction coefficient and keeping the feet near the ground act as the limiting factor given sufficiently powerful actuators. A parabolic relationship emerged for sprint time versus linkage lengths providing valuable insight into the parameters to use for the platform design. To the authors knowledge, no prior research has focused on rapid acceleration and braking manoeuvres of a biped in one optimisation problem, let alone providing insight for the physical bipedal robotic platform.

Copper microwire is an essential metal widely used in micro-electron mechanical systems. Since micro/nano material usually demonstrates unique mechanical properties due to scaling effect, copper microwire mechanical properties need to be investigated for better adhibition. Herein, we propose a nanorobotics manipulation system for copper microwire insitu twisting test. Firstly, a system with six degree-of-freedoms (DOFs) nanorobotic manipulator integrated inside scanning electron microscope (SEM) is introduced. Secondly, a positioning and assembly method for copper microwire specimen are proposed to solve the mismatching problem. Finally, the copper microwire is twisted in-situ and its properties are investigated and analyzed. The copper microwire sample fracture morphology shows a severe plastic deformation and being along with the emergence of deformation twin and intertwine, which exhibit strong scaling effects. This system provides a new method for in-situ twisting test, which paves the way for mechanical characterization inside SEM and benefits the fundamental nanomaterial research immensely.

This project seeks to generate small Unmanned Aerial System (sUAS) flight paths that are broadly understood by the general population and can communicate states about both the sUAS and its understanding of the world. Previous work in sUAS flight paths has sought to communicate intent, destination, or emotion of the system without focusing on concrete states (e.g., low battery, landing, etc.). This work leverages biologically-based flight paths and experimental methodologies from human-human and human-humanoid robot interactions to assess the understanding of avian flight paths to communicate sUAS states to novice users. If successful, this work should inform: the human-robot interaction community about the perception of flight paths, sUAS manufacturers on how their systems could communicate with both operators and bystanders, and end users on ways to communicate with others when flying systems in public spaces. General design implications and future directions of work are suggested to build on the results here, which suggest that novice users gravitate towards labels they understand (draw attention and landing) while avoiding more technical labels (lost sensor).

This paper introduces a new method for inverse reinforcement learning in large state spaces, where the learned reward function can be used to control high-dimensional robot systems and analyze complex human movement. To avoid solving the computationally expensive reinforcement learning problems in reward learning, we propose a function approximation method to ensure that the Bellman Optimality Equation always holds, and then estimate a function to maximize the likelihood of the observed motion. The time complexity of the proposed method is linearly proportional to the cardinality of the action set, thus it can handle large state spaces efficiently. We test the proposed method in a simulated environment on reward learning, and show that it is more accurate than existing methods and significantly better in scalability. We also show that the proposed method can extend many existing methods to large state spaces. We then apply the method to evaluating the effect of rehabilitative stimulations on patients with spinal cord injuries based on the observed patient motions.

Human-in-the-loop manipulation is useful in when autonomous grasping is not able to deal sufficiently well with corner cases or cannot operate fast enough. Using the teleoperator's hand as an input device can provide an intuitive control method but requires mapping between pose spaces which may not be similar. We propose a low-dimensional and continuous teleoperation subspace which can be used as an intermediary for mapping between different hand pose spaces. We present an algorithm to project between pose space and teleoperation subspace. We use a non-anthropomorphic robot to experimentally prove that it is possible for teleoperation subspaces to effectively and intuitively enable teleoperation. In experiments, novice users completed pick and place tasks significantly faster using teleoperation subspace mapping than they did using state of the art teleoperation methods.

Recent intuitive robot programming approaches operate on task level, enabling programmers to intuitively arrange or compose encapsulating robot capabilities (skills). This paper presents an approach to intuitively create (sub-)skills. General guidelines for assembly process descriptions #2860 provided by the Association of German Engineers VDI are applied to robot programming, in particular addressing assembly applications. The guidelines are exemplarily applied to the constraint-based approach iTaSC (instantaneous Task Specification using Constraints), presenting a procedure to hierarchically combine elementary processes to (sub-)skills. Six elementary processes are identified to be sufficient to implement a wide variety of assembly tasks. An iTaSC implementation was developed and two exemplarily assembly tasks were realized to evaluate the approach.

Endotracheal intubation is one of the most common procedures performed worldwide in emergency departments and operating rooms. It is a highly complicated procedure susceptible to failure. This paper presents a robotic prototype, called IntuBot, designed to automate this procedure. The hardware system consists of a stepper motor to steer the stylet in forward and backward motions and two servo motors to generate bending at the stylet tip to navigate through a patient's airway. A real-time vision-based navigation algorithm is also presented to guide the stylet to localize the vocal cords, which is the tubes ultimate target. For pre-clinical testing, we 3D printed and then molded a silicone model of the airway from the mouth to the vocal cords based on a series of actual CT scan images. The prototype was tested for its steering capabilities.

Pneumatic stepper motors are one of the promising alternative actuation methods for motion control in environments where electromagnetic (EM) motors cannot be used. Due to the lack of commercial off-the-shelf products, researchers working on MR compatible robotics have to develop their own pneumatic actuators. This imposes extensive costs and delays on the development process. Additionally, the current solutions are limited in their range of specifications and are difficult to manufacture. In this paper, proof-of-concept-prototypes for a family of parametrically designed, electromagnetically stealth, rotational pneumatic stepper motors are presented. The main objective of the paper is to demonstrate a general purpose non-electromagnetic actuation method, which can be customized and integrated into any design. Customizability, miniaturization, safety and affordability are some of the key features of the presented work. The developed prototypes are entirely 3D-printed and contain no sealing, bearing or lubrication. Thanks to the low production cost, the motor can be used as a disposable part in surgical applications. Experiments demonstrate effectiveness of the design in terms of cost-efficiency, versatility, MRI-compatibility, speed and performance. In order to optimize the design and control algorithm, empirical equations are presented describing response time of a pneumatic system to sequential pressure signals. A rotational speed of 800 rpm, total volume of 4.6 cm3 and resolution of 3° are some of the design attributes. The effects of clearance on stick-slip effect and leakage in a 3D printed cylinder-piston are also presented.

Comprehension of spoken natural language is an essential skill for robots to communicate with humans effectively. However, handling unconstrained spoken instructions is challenging due to (1) complex structures and the wide variety of expressions used in spoken language, and (2) inherent ambiguity of human instructions. In this paper, we propose the first comprehensive system for controlling robots with unconstrained spoken language, which is able to effectively resolve ambiguity in spoken instructions. Specifically, we integrate deep learning-based object detection together with natural language processing technologies to handle unconstrained spoken instructions, and propose a method for robots to resolve instruction ambiguity through dialogue. Through our experiments on both a simulated environment as well as a physical industrial robot arm, we demonstrate the ability of our system to understand natural instructions from human operators effectively, and show how higher success rates of the object picking task can be achieved through an interactive clarification process.

We present a novel Augmented Reality (AR) approach, through Microsoft HoloLens, to address the challenging problems of diagnosing, teaching, and patching interpretable knowledge of a robot. A Temporal And-Or graph (T-AOG) of opening bottles is learned from human demonstration and programmed to the robot. This representation yields a hierarchical structure that captures the compositional nature of the given task, which is highly interpretable for the users. By visualizing the knowledge structure represented by a T-AOG and the decision making process by parsing the T-AOG, the user can intuitively understand what the robot knows, supervise the robot's action planner, and monitor visually latent robot states (e.g., the force exerted during interactions). Given a new task, through such comprehensive visualizations of robot's inner functioning, users can quickly identify the reasons of failures, interactively teach the robot with a new action, and patch it to the current knowledge structure. In this way, the robot is capable of solving similar but new tasks only through minor modifications provided by the users interactively. This process demonstrates the interpretability of our knowledge representation and the effectiveness of the AR interface.