ICRA 2017 论文列表

In surface treatment processes like plasma spraying and spray painting of workpieces, protecting the uninvolved surface by applying masking tape is a common process. Due to the operation complexity for different geometries, such taping tasks depend on a lot of manual works, which is tedious and tiring. This paper introduces an automatic agile robotic system and the corresponding algorithm to do the surface taping. The automatic taping system consists of a 3D scanner for workpiece 3D model reconstruction, a taping end-effector which is mounted on a robot manipulator to handle the taping task, and a rotating platform that is used to hold the workpiece. The surface covering method and the taping path planning algorithms using the scanned model are introduced. With the implementation of the compliance mechanism, the force feedback and the tape cutting mechanism, the system is able to tape flat, cylindrical, freeform, and grooved surfaces. Experiments conducted on taping an engine inner liner shows that the surface can be covered with uniform taping overlap and very little wrinkle. The proposed system is a useful taping package for industrial applications such as workpiece repairing and surface protection, where surface treatments are involved.

Mapping of underwater environments is a critical task for a range of activities from monitoring coral reef habitats to surveying submerged archaeological sites. While recent advances in methods for terrestrial mapping can achieve dense 3D reconstructions of scenes in real-time, there remains the challenge of transferring these methods to the underwater domain due to characteristic effects on propagation of light through the water column that violate the brightness constancy constraint used in terrestrial techniques. Current state-of-the-art methods for underwater 3D reconstruction exploit a physical model of light propagation underwater to account for such range-dependent effects as scattering and attenuation; however, these methods necessitate careful calibration of attenuation coefficients required by the physical model, or rely on rough estimates of these coefficients from prior lab experiments. The main contribution of this paper is to develop a novel method to achieve simultaneous estimation of attenuation coefficients for color correction during structure recovery of an underwater scene by integrating this estimation directly into the bundle adjustment step, which performs non-linear optimization. To validate the proposed method, an artificial scene is submerged in a pure water tank and surveyed with a stereo camera platform to simulate an underwater robotic survey in a controlled environment. The target structure is imaged in air with an RGB-D sensor to provide ground truth structure and color, and a color calibration board is place in the scene for further reference. Results show that the proposed method can automatically estimate a water-column aware model for color correction of underwater images simultaneously to 3D reconstruction of the submerged scene.

A large number of novel path planning methods for a wide range of problems have been described in literature over the past few decades. These algorithms can often be configured using a set of parameters that greatly influence their performance. In a typical use case, these parameters are only very slightly tuned or even left untouched. Systematic approaches to tune parameters of path planning algorithms have been largely unexplored. At the same time, there is a rising interest in the planning and robotics communities regarding the real world application of these theoretically developed and simulation-tested planning algorithms. In this work, we propose the use of Sequential Model-based Algorithm Configuration (SMAC) tools to address these concerns. We show that it is possible to improve the performance of a planning algorithm for a specific problem without the need of in-depth knowledge of the algorithm itself. We compare five planners that see a lot of practical usage on three typical industrial pick-and-place tasks to demonstrate the effectiveness of the method.

Robotic swarms are distributed systems that exhibit global behaviors arising from local interactions between individual robots. Each robot can be programmed with several local control laws that can be activated depending on an operator's choice of global swarm behavior. While some simple behaviors (e.g. rendezvous) with guaranteed performance on known objectives under strict assumptions have been studied in the literature, real missions occur in uncontrolled environments with dynamically arising objectives and require combinations of behaviors. Given a library of swarm behaviors, a supervisory operator commanding the swarm must choose a sequence of behaviors to execute in order to accomplish a particular task during a mission composed of many dynamically arising tasks. In this paper, we formalize the problem of finding an optimal behavior sequence to maximize swarm performance on a complex task. Given the swarm behavior library, a set of decision time points and a performance criterion, we present an informed search algorithm that computes the maximum performance behavior sequence. The algorithm is proven to be optimal and complete. A relevant modification is presented that generates bounded suboptimal solutions more quickly. We apply the algorithm to a swarm navigation application and a dynamic area coverage application, demonstrating the utility of our algorithm even in situations where the behaviors in the library have not been designed for the task at hand.

We propose a novel method for generating test scenarios for a black box autonomous system that demonstrate critical transitions in its performance modes. In complex environments it is possible for an autonomous system to fail at its assigned mission even if it complies with requirements for all subsystems and throws no faults. This is particularly true when the autonomous system may have to choose between multiple exclusive objectives. The standard approach of testing robustness through fault detection is directly stimulating the system and detecting violations of the system requirements. Our approach differs by instead running the autonomous system through full missions in a simulated environment and measuring performance based on high-level mission criteria. The result is a method of searching for challenging scenarios for an autonomous system under test that exercise a variety of performance modes. We utilize adaptive sampling to intelligently search the state space for test scenarios which exist on the boundary between distinct performance modes. Additionally, using unsupervised clustering techniques we can group scenarios by their performance modes and sort them by those which are most effective at diagnosing changes in the autonomous system's behavior.

Visual attention is the cognitive process that allows humans to parse a large amount of sensory data by selecting relevant information and filtering out irrelevant stimuli. This papers develops a computational approach for visual attention in robots. We consider a Visual-Inertial Navigation (VIN) problem in which a robot needs to estimate its state using an on-board camera and an inertial sensor. The robot can allocate limited resources to VIN, due to time and energy constraints. Therefore, we answer the following question: under limited resources, what are the most relevant visual cues to maximize the performance of visual-inertial navigation? Our approach has four key features. First, it is task-driven, in that the selection of the visual cues is guided by a metric quantifying the task performance. Second, it exploits the notion of anticipation, since it uses a simplified model for forward-simulation of robot dynamics, predicting the utility of a set of visual cues over a time horizon. Third, it is efficient and easy to implement, since it leads to a greedy algorithm for the selection of the most relevant visual cues. Fourth, it provides formal performance guarantees: we leverage submodularity to prove that the greedy selection cannot be far from the optimal (combinatorial) selection. Simulations and real experiments on agile micro aerial vehicles show that our approach leads to dramatic improvements in the VIN performance. In the easy scenarios, our approach outperforms the state of the art in terms of localization errors. In the most challenging scenarios, it enables accurate visual-inertial navigation while the state of the art fails to track robot's motion during aggressive maneuvers.

We present a new 3D probabilistic occupancy map representation for robotics applications by relaxing the commonly-assumed constraint that space must be perfectly tessellated. We replace the regular structure of 3D grids with an unstructured collection of non-overlapping, equally-sized spheres, which we call “atoms”. Abandoning the grid structure allows a more accurate representation of space directly tangent to surfaces, which facilitates a number of applications such as high fidelity surface reconstruction and surface-guided path planning. Maps composed of atoms can distinguish between free, occupied, and unknown space, support computationally efficient insertions and collision queries, provide free space planning guarantees, and achieve state-of-the-art memory efficiency over large volumes. This is achieved while simultaneously reducing quantization effects in the vicinity of surfaces and defining a useful implicit surface representation.

Robotic systems that are soft or incorporate soft actuators are well suited for operation in unstructured environments and for safe interactions with fragile objects. The majority, however, are tethered or burdened with bulky payloads of pumps and compressors. In a recent article we presented a sealed, inflated actuator composed of fluidically connected membrane dielectric elastomer actuators capable of large, repeatable, and stable deformations. Each membrane could switch between two identical volumes, and maintain its shape when an applied voltage was removed. Here we extend our previous work by simulating and demonstrating asymmetric stable deformations. In an experimental two-membrane setup, the membranes experience large and significantly different area strains of >100% and >550% when transitioning between stable states. With the addition of more membranes, this asymmetry can increase the number of discrete stable membrane sizes, allowing more complex control when later implemented in a mechanism.

In this paper we present an architecture for the operation of an assistive robot finally aimed at allowing users with severe motion disabilities to perform manipulation tasks that may help in daily-life operations. The robotic system, based on a lightweight robot manipulator, receives high level commands from the user through a Brain-Computer Interface based on P300 paradigm. The motion of the manipulator is controlled relying on a closed loop inverse kinematic algorithm that simultaneously manages multiple set-based and equality-based tasks. The software architecture is developed relying on widely used frameworks to operate BCIs and robots (namely, BCI2000 for the operation of the BCI and ROS for the control of the manipulator) integrating control, perception and communication modules developed for the application at hand. Preliminary experiments have been conducted to show the potentialities of the developed architecture.

We present a shared control method of painting 3D geometries, using a handheld robot which has a single autonomously controlled degree of freedom. The user scans the robot near to the desired painting location, the single movement axis moves the spray head to achieve the required paint distribution. A simultaneous simulation of the spraying procedure is performed, giving an open loop approximation of the current state of the painting. An online prediction of the best path for the spray nozzle actuation is calculated in a receding horizon fashion. This is calculated by producing a map of the paint required in the 2D space defined by nozzle position on the gantry and the time into the future. A directed graph then extracts its edge weights from this paint density map and Dijkstra's algorithm is then used to find the candidate for the most effective path. Due to the heavy parallelisation of this approach and the majority of the calculations taking place on a GPU we can run the prediction loop in 32.6ms for a prediction horizon of 1 second, this approach is computationally efficient, outperforming a greedy algorithm. The path chosen by the proposed method on average chooses a path in the top 15% of all paths as calculated by exhaustive testing. This approach enables development of real time path planning for assisted spray painting onto complicated 3D geometries. This method could be applied to applications such as assistive painting for people with disabilities, or accurate placement of liquid when large scale positioning of the head is too expensive.

The problem of optimal motion planing and control is fundamental in robotics. However, this problem is intractable for continuous-time stochastic systems in general and the solution is difficult to approximate if non-instantaneous nonlinear performance indices are present. In this work, we provide an efficient algorithm, PIPC (Probabilistic Inference for Planning and Control), that yields approximately optimal policies with arbitrary higher-order nonlinear performance indices. Using probabilistic inference and a Gaussian process representation of trajectories, PIPC exploits the underlying sparsity of the problem such that its complexity scales linearly in the number of nonlinear factors. We demonstrate the capabilities of our algorithm in a receding horizon setting with multiple systems in simulation.

This study presents a novel apprenticeship learning method to enable a learner to utilize demonstrations observed in an incompatible feature space. It is assumed that an expert and a learner follow non-identical Markov decision processes (MDPs), and a mapping function is estimated to obtain feature expectation of the demonstrations in an agent space. A conditional density estimation technique is used to represent the feature expectation in closed-form. The proposed method is useful because it is expected to alleviate intractable processes to explicitly specify correspondence of heterogeneous MDPs for apprenticeship learning. Additionally, the method does not require any sampling method to approximate integrals over an agent feature space. A simulation is used to demonstrate the validity of the proposed method in three domains in which it is not possible to directly compare the features of the expert and learner.

In visual SLAM, there are many software and hardware parameters, such as algorithmic thresholds and GPU frequency, that need to be tuned; however, this tuning should also take into account the structure and motion of the camera. In this paper, we determine the complexity of the structure and motion with a few parameters calculated using information theory. Depending on this complexity and the desired performance metrics, suitable parameters are explored and determined. Additionally, based on the proposed structure and motion parameters, several applications are presented, including a novel active SLAM approach which guides the camera in such a way that the SLAM algorithm achieves the desired performance metrics. Real-world and simulated experimental results demonstrate the effectiveness of the proposed design space and its applications.

In the field of aerial robotics, one of the key challenges is to enable aerial manipulators to exert substantial forces on the environment. Enabling this will allow the technology to perform meaningful tasks airborne, such as cleaning or grinding surfaces. While in contact and applying a large, continuous force, control of the UAV's attitude is a challenge. In this work, we show that a regular (PID-based) attitude controller is incapable of stabilizing aerial manipulators that apply physical contact forces on the environment that are comparable to the UAV's weight. We present a novel control algorithm that uses an LQR-optimized state feedback on the roll and yaw angle while in contact. Experiments on a UAV of 1.5 kg show that the proposed controller is capable of applying a contact force of over 15 N - equal to the UAV's weight - sustained for several minutes.

Humans utilize their torsos and arms while running to compensate for the angular momentum generated by the lower-body movement during the flight phase. To enable this capability in a humanoid robot, the robot should have human-like mass, a center of mass position, and inertial moment of each link. To mimic this characteristic, we developed an angular momentum control method using a humanoid upper body based on human motion. In this method, the angular momentum generated by the movement of the humanoid lower body is calculated, and the torso and arm motions are calculated to compensate for the angular momentum of the lower body. We additionally developed the humanoid upper-body mechanism that mimics the human link length and mass property by using carbon fiber reinforced plastic and a symmetric structure. As a result, the developed humanoid robot could generate almost the same angular momentum as that of human through human-like running motion. Furthermore, when suspended in midair, the humanoid robot produced the angular momentum compensation in the yaw direction.

This paper addresses the task of designing a modular neural network architecture that jointly solves different tasks. As an example we use the tasks of depth estimation and semantic segmentation given a single RGB image. The main focus of this work is to analyze the cross-modality influence between depth and semantic prediction maps on their joint refinement. While most of the previous works solely focus on measuring improvements in accuracy, we propose a way to quantify the cross-modality influence. We show that there is a relationship between final accuracy and cross-modality influence, although not a simple linear one. Hence a larger cross-modality influence does not necessarily translate into an improved accuracy. We find that a beneficial balance between the cross-modality influences can be achieved by network architecture and conjecture that this relationship can be utilized to understand different network design choices. Towards this end we propose a Convolutional Neural Network (CNN) architecture that fuses the state-of-the-art results for depth estimation and semantic labeling. By balancing the cross-modality influences between depth and semantic prediction, we achieve improved results for both tasks using the NYU-Depth v2 benchmark.

The mechanical design of robotic hands has been converging towards low-inertia, tendon-driven strategies. As tendon driven robotic fingers are serial chain systems, routing strategies with compliant tendons lead to multi-articular coupling between the degrees of freedom. We propose a generalized analysis of such serial chain linkages with coupled passive joint stiffnesses. We analyze the effect of such coupling on maximum achievable stiffness control boundaries while maintaining passivity at the actuators by analytically deriving the boundaries. We believe that we can use this information to form mechanical design guidelines for intelligently selecting arrangements of compliance elements (mechanical springs) and transmission strategies, i.e. tendon routing and pulley radii, to provide intrinsic stability and customizable controller stiffness limits for high performance manipulation in robotic hands.

This paper introduces an analytical approach for the design of variable stiffness mechanisms. The basis of this approach is a general model - representing the potential energy function and the physical constraints - covering the design space of variable stiffness mechanisms. Using this model, we present a systematic procedure to analytically define classes of variable stiffness mechanisms from first principles. Consequently, we identify mechanisms capable of infinite range stiffness modulation using bounded motor forces, and define the simplest mathematical model representing mechanisms in this class. A prototype mechanism consistent with this canonical model is designed, fabricated and experimentally tested. The experimental data are consistent with our theoretical predictions showing constant motor force independent of the output deflection and output stiffness when the mechanism is subject to external load.

By bonding micron-thin polymer layers with an electrically conductive layer, we can utilize the bi-metal effect to create a soft, thin, electrically-activated thermal actuator (ETA), which can be used in printed soft robots. Using ETAs have shown quite remarkable potential in previous works. However, to reveal their true capabilities, we propose to utilize a multi-layered ETA. In this research, we analytically model the actuator for both the bi-, and more importantly, tri-layer ETAs and analyze their behavior. We develop a simple intuitive model to better explain the benefits of the tri-layer actuation and to estimate the optimal design parameters. We verify the models by conducting an experiment to measure the curvature change for different third layer thicknesses. The experiments show a significant improvement in performance of the tri-layer actuators and display a good fit between the simplified analytic model and the full model.