IROS 2015 论文列表

Agile robot locomotion on rough terrain is highly dependent on the ability to perceive the environment. In this paper, we show how the interaction between a reactive control framework and an online mapping system can significantly improve the trotting performance on irregular terrain. In particular, this new locomotion controller increases the stability of the robot and reduces frontal leg and shin collisions with obstacles by correcting in realtime the foothold locations. The mapping system uses an RGB-D sensor and a motion capture system to build a three dimensional map of the surroundings of the robot. While the robot is trotting, the control framework requests in advance a local heightmap around the next nominal foothold position. Then, an optimized foot placement location is estimated by applying visual pattern classification on the acquired heightmaps, and the leg endpoint trajectory is modified accordingly. The foothold correction is performed independently for each leg. To show the effectiveness of our approach the controller was tested both in simulation and experimentally with our 80 kg hydraulic quadruped robot, HyQ. The results show that visual based reaction through pattern classification is a promising approach to increase locomotion robustness over challenging terrain.

Vehicle localization in large-scale urban environments has been commonly addressed as a map-matching problem in the literature. Generally, the maps are 2D images of the world where each pixel covers a part of it. However, building maps for large-scale urban environments requires driving the vehicle along the desired path at least once. In order to simplify this task, in this work, we propose a new localization system that uses satellite aerial map-images available on the Internet to localize a vehicle in a complex urban environment. Satellite aerial map-images are compared against re-emission maps built from the infrared reflectance information of the vehicle's LiDAR. Normalized Mutual Information (NMI) is used to compare re-emission and aerial map images. A Particle Filter Localization strategy is applied for vehicle's localization. As a result, the system has an accuracy of 0.89m in a test course with 6.5km. Our system can be used continuously without losing track, and it works even in dark and partially occluded areas.

Additive manufacturing is a technology in constant evolution. It is able to produce objects otherwise either impractical or unfavorable for traditional manufacturing technologies, especially in relatively limited quantities. The advancement of Selective Laser Melting (SLM) has made it possible to manufacture functional, production-quality components directly from rapid prototyping for long-term use. This work studies the SLM process and identifies the optimal process parameters for the rapid manufacturing of a miniaturized robotic surgical instrument. The proposed robotic instrument has been designed to exploit the advantages of rapid manufacturing and rapid assembly, following a shift from large scale manufacturing of generalized surgical instruments to the production of small batches of patient or procedure-specific.

Robotics applications are characterized by a huge amount of variability. Their design requires the developers to choose between several variants, which relate to both functionalities and hardware. Some of these choices can be taken at deployment-time, however others should be taken at run-time, when more information about the context is known. To make this possible, a software system needs to be able to reason about its current state and to adapt its architecture to provide the configuration that best suites the context. This paper presents a model-based approach for run-time adaptation of robotic systems. It defines a set of orthogonal models that represent the system architecture, its variability, and the state of the context. Additionally it introduces a set of algorithms that reason about the knowledge represented in our models to resolve the run-time variability and to adapt the system architecture. The paper discusses and evaluates the approach by means of two case studies.

This work presents a flexible system to reconstruct 3D models of objects captured with an RGB-D sensor. A major advantage of the method is that unlike other modelling tools, our reconstruction pipeline allows the user to acquire a full 3D model of the object. This is achieved by acquiring several partial 3D models in different sessions—each individual session presenting the object of interest in different configurations that reveal occluded parts of the object — that are automatically merged together to reconstruct a full 3D model. In addition, the 3D models acquired by our system can be directly used by state-of-the-art object instance recognition and object tracking modules, providing object-perception capabilities to complex applications requiring these functionalities (e.g. human-object interaction analysis, robot grasping, etc.). The system does not impose constraints in the appearance of objects (textured, untextured) nor in the modelling setup (moving camera with static object or turn-table setups with static camera). The proposed reconstruction system has been used to model a large number of objects resulting in metrically accurate and visually appealing 3D models.

This paper presents a new mobile robotic platform (R-Mo) which can reduce unexpected variations in height as well as pitch angle of its main body while traversing rough terrains. As a measure for the smooth movement of mobile platform, the variations in height and pitch angle are chosen in this study. Then, the kinematic analysis on the Rocker-Bogie mechanism is carried out to investigate its variations in height and pitch angle on rough terrains. Based on this result, a new mobile platform is systematically designed by combining the Rocker-Bogie with the inverse four bar linkage. The extensive experiments are carried out by using the Rocker-Bogie mechanism and the proposed R-Mo against rough terrain, which validate that in comparison with the Rocker-Bogie mechanism, the average and maximum variations in height of proposed R-Mo are reduced by 12.72% and 5.96%, respectively, and the average and maximum variations in pitch angle of proposed R-Mo are considerably reduced by 65.87 % and 60.53 %, respectively.

In this paper we introduce an index for the quantification of anthropomorphism of robot arms. The index is defined as a weighted sum of specific metrics which evaluate the similarities between the human and robot arm workspaces, providing a normalized score between 0 (non-anthropomorphic artifacts) and 1 (human-identical artifacts). The human arm workspaces were extracted using data reported in anthropometry studies. The formulation is general enough to allow utilization in various applications, by adjusting the weighting factors according to the specifications of each study. The proposed methodology can be used for assessing the human-likeness of existing robot arms as well as to provide specifications for the design of new anthropomorphic robots and prosthetic devices. To assess the efficiency of the proposed methods a comparative analysis between five kinematically different robot arm models is conducted and simulated paradigms are presented.

An objective of an autonomous sociable robot is to meet the needs and preferences of a human user. However, this can sometimes be at the expense of the robot's own ability to understand social signals produced by the user. In particular, human preferences of distance (proxemics) to the robot can have significant impact on the performance rates of its automated speech and gesture recognition systems. In this work, we investigated how people perceive a sociable robot based on its performance at different locations. We performed an experiment in which a robot's ability to understand social signals was artificially attenuated across distance; robot maximum, minimum, and average performance rates were also varied. Participants (N = 160) instructed a robot using speech and pointing gestures, and then responded to a questionnaire to provide their subjective evaluations of the robot.We identified significant relationships between robot performance and user perceptions of robot competence, anthropomorphism, engagement, likability, and technology adoption; we identified no relationship between human-robot distance and subjective measures, which contrasts related work. Our results have significant implications for autonomous sociable robot design.

We control the motion of helical microrobots with average diameter of 500 µm in two-dimensional (2D) and three-dimensional (3D) spaces using two synchronized rotating dipole fields. The utilization of the two synchronized dipole fields not only increases the magnetic torque exerted on the magnetic dipole of the helical microrobot but also eliminates the magnetic field gradients along its lateral directions. Our finite element simulations and experimental results show that the utilization of two rotating dipole fields increases the magnetic field by 100%, as opposed to single rotating magnetic field. In addition, we show that the magnetic field gradient within the workspace of the microrobot is eliminated. Therefore, the lateral oscillations of the helical microrobot are mitigated within the center of two rotating dipole fields, and hence the motion of the microrobot is stabilized inside tubes with relatively large inner diameters, as opposed to the diameter of the helical microrobot. This strategy allows the microrobot to compensate for gravity and swim in 3D space inside water reservoirs at an average speed of 0.25 body lengths per second. In addition, closed-loop motion control of the helical microrobot is achieved in 2D space at an average speed of 2 mm/s and maximum steady-state error of 100 µm.

Unmanned aerial vehicles (UAVs) such as multicopters are expected to be used for inspection of aged infrastructure or for searching damaged buildings in the event of a disaster. However, in a confined space in such environments, UAVs suffer a high risk of falling as a result of contact with an obstacle.

This paper presents a non-minimal order dynamics model for many analysis, simulation, and control problems of constrained mechanical systems passing through singular configurations during their motion by making use of linear projection operator. The distinct features of this model describing dynamics of the dependent coordinates are: i) The mass matrix M̅(q) is always positive definite even at singular configurations; ii) matrix equation is skew symmetric, where all nonlinear terms are lumped into vector C̅(q, q̇)q̇ after elimination of constraint forces. Eigenvalue analysis shows that the condition number of the constraint mass matrix can be minimized upon adequate selection of a scalar parameter called “virtual mass” thereby reducing the sensitivity to round-off errors in numerical computation. It follows by derivation of two oblique projection matrices for computation of constraint forces and actuation forces. It is shown that projection-based model allows feedback control of dependent coordinates which, unlike reduced-order dependent coordinates, uniquely define spatial configuration of constrained systems.

This paper addresses the problem of constrained motion for a manipulator performing a task while in contact with the environment, and investigates two force control frameworks, one based on projected inverse dynamics, and one based on optimal control. Firstly, we propose a control method based on projected inverse dynamics, which directly exploits the contact constraints to minimise the instantaneous joint torques needed to perform a task. Secondly, we propose an optimal control strategy which provides a tool to minimise the joint torques over an interval of time. We show how contact constraints can be used as optimisation constraints in the definition of the problem, and how to formulate the optimal control problem directly using projected dynamics. Initially we explore a positional control problem, where the robot is required to follow a desired path, and show that both of the proposed methods can satisfy the positional task while significantly reducing the joint torques as compared to simple kinematic control and also classical inverse dynamics control. We also show that the proposed optimal control method outperforms the pure projected inverse dynamics method in terms of minimising the required joint torques. We then show how each method can be extended to follow a desired path while also exerting a desired contact force. Again, the method incorporating optimal control is shown to satisfy the task requirements with significantly smaller commanded torques than the pure projected inverse dynamics method. To confirm the analysis, and demonstrate proof of concept, we present the results of empirical experiments with a simulated 3-degree-of-freedom planar manipulator which is constrained to move while in contact with a rigid surface.

Tendon-Driven Manipulators (TDM) have been used in various applications requiring low inertia robots having multiple degrees-of-freedom (DOF) and redundancy. TDMs are generally actuated by gear electric motors placed at the base of the robot, and consequently require complex cable tension feedback from force sensors or dynamic modelling to maintain cables under tension. This paper presents a high performance TDM actuated by magneto-rheological clutches along with a specialized motion control algorithm, termed Projected PID, that requires no tension feedback. A 2-DOF proof-of-concept TDM powered by magneto-rheological clutches is used to demonstrate overall controller performance, and a reconfigurable 2-DOF TDM powered by direct-drive electric motors is used to demonstrate the controller's ability to compensate for configuration variations and actuator failure. Experimental results also demonstrate the ability of Projected PID to control magneto-rheological cable-driven TDMs with high accuracy.

This paper discusses the application of a procedural semantics framework to autonomous robots. We model insights from decades of research in cognitive linguistics and empirical research to develop a system capable of autonomously producing and interpreting German locative utterances. The system is tested in the real world using a population of robots that talk to each other via natural language.

A high machining precision is the most important goal for the fabrication of three dimensional nanometer scale structure. The force-based atomic force microscope (AFM) can reach a very high precision in the structure machining, but the machining depth is difficult to predict for different sample material or machining parameters. The nano machining system in this paper adopts traditional displacement-based machining method, aiming at a few hundred nanometers machining depth. The machining tool (a probe) is kept motionless during the structure machining, so the structure size is determined by a high precise x/y/z three degree of freedoms (DOF) stage. A probe suspension with bridge-type amplification mechanism is designed and optimized, through the theoretical analysis and finite element analysis (FEA), the amplification factor, Z-direction stiffness and first natural frequency of the probe suspension are about 5, 5000N/m, 130HZ, respectively.

This paper focuses on a surveillance mission performed by an autonomous mobile robot in an environment. In this mission, the robot is required to detect as many intruders as possible by utilizing the mobility and monitoring capability. However, the robot does not know where and how many intruders are in the environment. Without the information, it is impossible for the robot to estimate an intrusion trend and detect the unknown intruders. For this challenge, we propose a novel surveillance method in order for the robot to estimate the intrusion trend. For this purpose, Bayes' rule is basically used. This probabilistic approach allows the robot to express the intruders through existence probability and appearance frequency. As the contribution of this paper, the robot is enabled to mainly monitor the places with the intrusion trend, in addition to other minor places. Through simulation experiments, the effectiveness of the proposed method for the unknown intruders is shown. Furthermore, an adaptability of the robot based on the probabilistic approach to various kinds of intrusion trends is discussed.

Classical approaches towards learning coordinated movement tasks often represent a movement in a sequential and exclusive fashion. Introducing concurrency allows to decompose such tasks into a number of separate sequences, for instance for two different end-effectors. While this results in a compact and generic representation of the individual movement primitives (MPs), it is a hard problem to learn their temporal and causal organization. This paper presents a concept for learning movement tasks that require the coordination of several controlled effectors of a robot. We firstly introduce a concept to learn and estimate the progress of individual MPs from a low number of demonstrations. Secondly, we propose a representation of the task that incorporates several concurrent sequences of MPs. Combining these two elements allows to learn and reproduce coordinated bi-manual movement tasks robustly. The synchronization of the concurrent MPs is achieved implicitly using the progress prediction. The approach is evaluated in two simulation studies with a 25 degrees of freedom two-arm robot performing a pick-and-place task.

In robot programming by demonstration (PBD) for assembly tasks, one of the important topics is to inference the poses and spatial relations of parts during the demonstration. In this paper, we propose a world model called assembly graph (AG) to achieve this task. The model is able to represent the poses of all parts, the relations, observations provided by vision techniques and prior knowledge in a unified probabilistic graph. Then the problem is stated as likelihood maximization estimation of pose parameters with the relations being the latent variables. Classification expectation maximization algorithm (CEM) is employed to solve the model. Besides, the contradiction between relations is incorporated as prior knowledge to better shape the posterior, thus guiding the algorithm find a more accurate solution. In experiments, both simulated and real world datasets are applied to evaluate the performance of our proposed method. The experimental results show that the AG gives better accuracy than the relations as deterministic variables (RDV) employed in some previous works due to the robustness and global consistency. Finally, the solution is implemented into a PBD system with ABB industrial robotic arm simulator as the execution stage, succeeding in real world captured assembly tasks.

Self-folding is a promising technique for assembling robots from flat sheets. However, existing implementations do not include reliable methods for sensing the folding angle, making feedback control impossible. In this paper, we present novel angle sensors for foldable robots and machines. They are inkjet printed and fully integrated into robots' laminate. This additional sensor layer tracks the angle motion of robot hinges, to better guide robot assembling by folding and to perform more complicated tasks that requires feedback control, making folded robots more capable in real world applications. We introduce the fabrication process, property assessments, and demonstrate sensor performance by measuring folding angles of a cube and controlling folds on a gripper.