ICRA 2017 论文列表

This paper presents a new approach for error robust assembly sequence planning which uses haptic rendering models (HRMs) for the representation of assemblies. Our assembly planning system uses HRMs for collision test along mating vectors, which are generated by stereographic projection. The planner stores the vectors in 2 1/2D distance maps providing fast and efficient access for the later evaluation while AND/OR-graphs contain possible sequences. Haptic rendering models facilitate the processing compared to faulty triangle meshes providing fast and geometry independent collision tests as colliding parts can easily be identified and handled accordingly. In addition, part and material related properties can be annotated. We present a fast and simple approach handling approximation inconsistencies, which occur due to discretization errors, based only on the properties of the haptic rendering models. The paper concludes with feasible results for various assemblies and detailed calculation times underlining the effectiveness of our approach.

This paper presents that the joint torque control capability can be enhanced by adding physical damper to a series elastic actuator (SEA). Joint torque tracking of standard SEA has known limitations that the torque dynamics has an relative order of two, and, as a consequence, the torque controller often requires acceleration feedback when the desired torque is defined by a function of velocity (for example, compliance control). This limitation can be removed by introducing physical damping, reducing the relative degree of torque dynamics by one. Based on this observation, we design a robust controller using the disturbance observer technique. The resulting control law is given by a feed-forward term combined with PI control. The proposed controller is verified in simulation and experiment.

A fundamental problem in human-robot collaboration is to ensure safety for humans being located in the workspace of the robot. Several new robots, referred to as collaborative robots, are pushing into the market. Most of these so-called co-bots have similar properties. They are small, lightweight and designed with big roundings to ensure safety in the case of a collision with a human. Equipped with torque sensors, external torque observers, tactile skins, etc., they are able to stop the robot when an emergency occurs. While developing more and more co-bots, the main focus lies on the robot itself. But to make a robot deployable, a special tool for a defined task is needed. These tools are often sharp-edged and dangerous in case of a collision with a human. In this paper we present a new safety module for robots to ensure safety for different tools in collaborative tasks. This module, filled with air pressure during the robot motion, covers mounted tools and carried workpieces. In case of a non or very slow moving robot, the safety module is able to pull back and the tool is uncovered. In our experiments we found out that we can increase the velocity up to 1 m/s while satisfying the requirements of the ISO/TS 15066 and retain the full functionality of the tool.

This work introduces a wearable system to provide situational awareness for blind and visually impaired people. The system includes a camera, an embedded computer and a haptic device to provide feedback when an obstacle is detected. The system uses techniques from computer vision and motion planning to (1) identify walkable space; (2) plan step-by-step a safe motion trajectory in the space, and (3) recognize and locate certain types of objects, for example the location of an empty chair. These descriptions are communicated to the person wearing the device through vibrations. We present results from user studies with low- and high-level tasks, including walking through a maze without collisions, locating a chair, and walking through a crowded environment while avoiding people.

We present a data processing pipeline to online estimate ego-motion and build a map of the traversed environment, leveraging data from a 3D laser, a camera, and an IMU. Different from traditional methods that use a Kalman filter or factor-graph optimization, the proposed method employs a sequential, multi-layer processing pipeline, solving for motion from coarse to fine. The resulting system enables high-frequency, low-latency ego-motion estimation, along with dense, accurate 3D map registration. Further, the system is capable of handling sensor degradation by automatic reconfiguration bypassing failure modules. Therefore, it can operate in the presence of highly dynamic motion as well as in dark, texture-less, and structure-less environments. During experiments, the system demonstrates 0.22% of relative position drift over 9.3km of navigation and robustness w.r.t aggressive motion such as highway speed driving (up to 33m/s).

Robot Reinforcement Learning (RL) algorithms return a policy that maximizes a global cumulative reward signal but typically do not create diverse behaviors. Hence, the policy will typically only capture a single solution of a task. However, many motor tasks have a large variety of solutions and the knowledge about these solutions can have several advantages. For example, in an adversarial setting such as robot table tennis, the lack of diversity renders the behavior predictable and hence easy to counter for the opponent. In an interactive setting such as learning from human feedback, an emphasis on diversity gives the human more opportunity for guiding the robot and to avoid the latter to be stuck in local optima of the task. In order to increase diversity of the learned behaviors, we leverage prior work on intrinsic motivation and empowerment. We derive a new intrinsic motivation signal by enriching the description of a task with an outcome space, representing interesting aspects of a sensorimotor stream. For example, in table tennis, the outcome space could be given by the return position and return ball speed. The intrinsic motivation is now given by the diversity of future outcomes, a concept also known as empowerment. We derive a new policy search algorithm that maximizes a trade-off between the extrinsic reward and this intrinsic motivation criterion. Experiments on a planar reaching task and simulated robot table tennis demonstrate that our algorithm can learn a diverse set of behaviors within the area of interest of the tasks.

We document empirically stable bounding using an actively powered spine on the Inu quadrupedal robot, and propose a reduced-order model to capture the dynamics associated with this additional, actuated spine degree of freedom. This model is sufficiently accurate as to roughly describe the robots mass center trajectory during a bounding limit cycle, thus making it a potential option for low dimensional representations of spine actuation in steady-state legged locomotion.

In this paper we evaluate the predictive performance of six commonly used rigid body impact models on real planar impacts captured with a motion tracking system. We propose a metric to evaluate the performance of impact models on a task (based on predicting post impact momentum) and use this metric to tune the six parametric models. We evaluate model performance in predicting impact outcomes against the defined metric and discuss the implications of uncertainty in geometric models and initial conditions. We show that the models can fairly effectively predict the outcomes of single impacts on our chosen task. We motivate further study into consensus and hybrid impact models by showing that a hypothetical hybrid model would significantly outperform the isolated models by providing a post-hoc model that demonstrates an upper bound on the combined predictive power of the models. We use perturbation analysis to compute the predictive range of the models and show that bifurcations can cause the model predictions to cluster into regions of the state space.

The rapid advancement of robotics poses the problem of a deep integration of robotic systems in human environments. In order to achieve this symbiosis between humans and robots, the artificial systems have to take into account one of the most important aspects in human life: emotions. The recognition and understanding of human emotions is crucial for robotic systems to behave in appropriate ways according to the situation and smoothly integrate with all the different aspects of human life. This paper proposes a novel algorithm which uses state-of-the-art techniques in Machine Learning, in particular Recurrent Neural Networks, to automatically infer emotional clues from non-stylized motions (i.e. motions which are not supposed to convey emotional information as primary goal). This algorithm recognized human emotions with an accuracy between 0.68 and 0.80, depending on the considered motion, and clearly overcomes human capacity in the same task for the considered cases studied. Since the implemented algorithm is able to perform online, its results can be used to allow a behavioural programming which gives the robot the flexibility to act in a more human-oriented way.

Road detection from the perspective of moving vehicles is a challenging issue in autonomous driving. Recently, many deep learning methods spring up for this task because they can extract high-level local features to find road regions from raw RGB data, such as Convolutional Neural Networks (CNN) and Fully Convolutional Networks (FCN). However, how to detect the boundary of road accurately is still an intractable problem. In this paper, we propose a siamesed fully convolutional network (named as “s-FCN-loc”) based on VGG-net architecture, which is able to consider RGB-channel, semantic contour and location prior simultaneously to segment road region elaborately. To be specific, the s-FCN-loc has two streams to process original RGB images and contour maps respectively. At the same time, the location prior is directly appended to the last feature map to promote the final detection performance. Experiments demonstrate that the proposed s-FCN-loc can learn more discriminative features of road boundaries and converge 30% faster than the original FCN during the training stage. Finally, the proposed approach is evaluated on KITTI road detection benchmark, and achieves a competitive result.

Dense depth map estimation from stereo cameras has many applications in robotic vision, e.g., obstacle detection, especially when performed in real-time. The range in which depth values can be accurately estimated is usually limited for two-camera stereo setups due to the fixed baseline between the cameras. In addition, two-camera setups suffer from wrong depth estimates caused by local minima in the matching cost functions. Both problems can be alleviated by adding more cameras as this creates multiple baselines of different lengths and since multi-image matching leads to unique minima. However, using more cameras usually comes at an increase in run-time. In this paper, we present a novel embedded system for multi-baseline stereo. By exploiting the parallelization capabilities within FPGAs, we are able to estimate a depth map from multiple cameras in real-time. We show that our approach requires only little more power and weight compared to a two-camera stereo system. At the same time, we show that our system produces significantly better depth maps and is able to handle occlusion of some cameras, resulting in the redundancy typically desired for autonomous vehicles. Our system is small in size and leight-weight and can be employed even on a MAV platform with very strict power, weight, and size requirements.

The integrated fabrication of body structures, actuators, sensors, and electronic circuits into one robot system is an open problem in robotics. Simple and rapid construction of electric actuators in the body through existing approaches is difficult. We take advantage of the liquid-to-gas phase change, and propose an electric phase-change actuator comprising a printable fluidic actuator controlled by an inkjet printed electric heater. The actuator can easily be integrated with origami robots. We theoretically analyze the dynamics of electro-fluidic conversion in the actuator and compare it with actual measurement data. The proposed actuator is verified in real examples of a shape-shifting origami structure and a robot gripper with a printed touch sensor.

This study proposes a multi-level trajectory analysis method for modeling traffic behavior from an ego-centric view, where on-road vehicle trajectories are collected based on the authors' previous studies of an on-board system consisting of multiple 2D lidar sensors. From an input set of trajectories, a set of hot regions (topics) that trajectory points most frequently present are first discovered using a sticky HDP-HMM; then, the major paths of the trajectories' transitions across different hot regions are extracted by recursively mining frequent subsequences of topics; and finally, paths are modeled using a hierarchical hidden Markov model (HHMM), where the intra-path dynamics is represented using an HMM, in which each state corresponds to a hot region, while the inter-path transition is assumed to be Markovian. The model could be used for behavior prediction, i.e. whenever a vehicle is detected in a scene, predicting which route it will probably follow and how its trajectory will probably develop over time, which is essential to interpreting the potential risks for longer time horizons. Experiments are conducted using a large set of vehicle trajectories collected from motorways in Beijing, and promising results are presented.

Compliant actuators have found their place in areas of prosthetics, rehabilitation and robot locomotion because they enable safe human-robot and stable robot-environment interaction, both non-trivial to achieve using conventional rigid actuation. These actuators are capable of varying their equilibrium position and apparent output stiffness in a way humans change the resting position and compliance of their limbs. Just like antagonistically actuated human joints, these actuators require two motor units to provide control over the equilibrium position and the positive joint stiffness. Here we present a novel compliant actuation concept which affords control over the equilibrium position and joint stiffness using a single motor unit. In order to achieve this unconventional functionality, the actuator combines a passive positive feedback (negative stiffness) mechanism with an efficiently tunable negative feedback (positive stiffness) mechanism. This provides a novel design with two distinct operation modes, one leading to unprecedented stiffness tunability, while the other enabling equilibrium point controllability. We present the first practical implementation of this actuator using a prototype prosthetic limb design along with experimental data testifying the range of tunability, covering compliant to rigid behaviour, without paying much on the power input.

We present new algorithms to perform fast probabilistic collision queries between convex as well as non-convex objects. Our approach is applicable to general shapes, where one or more objects are represented using Gaussian probability distributions. We present a fast new algorithm for a pair of convex objects, and extend the approach to non-convex models using hierarchical representations. We highlight the performance of our algorithms with various convex and non-convex shapes on complex synthetic benchmarks and trajectory planning benchmarks for a 7-DOF Fetch robot arm.

Efficient navigation planning for mobile robots in complex environments is a challenging problem. In this paper we consider the path planning problem for mobile robots with adjustable relative wheel positions, which further increase the navigation capabilities. In particular we account for changes of these relative wheel positions during planning time, thus fully leveraging the capabilities of the robot. Whereas these additional degrees of freedom increase flexibility, they introduce a more challenging planning problem. The approach proposed in this paper is built upon a search-based planner. We describe how to flexibly integrate joint angle changes in the path planning process and furthermore propose a representation of the robot configuration that substantially reduces the computational burden. In addition, we introduce search guidance heuristics that are particularly useful in environments in which a robot is required to pass over obstacles, such as on agricultural fields. An extensive evaluation on simulated and real-world data with our BoniRob agricultural robot demonstrates the efficiency of our approach.

We present a novel approach for collision-free global navigation for continuous-time multi-agent systems with general linear dynamics. Our approach is general and can be used to perform collision-free navigation in 2D and 3D workspaces with narrow passages and crowded regions. As part of pre-computation, we compute multiple bridges in the narrow or tight regions in the workspace using kinodynamic RRT algorithms. Our bridge has certain geometric properties that enable us to calculate a collision-free trajectory for each agent using simple interpolation at runtime. Moreover, we combine interpolated bridge trajectories with local multi-agent navigation algorithms to compute global collision-free paths for each agent. The overall approach combines the performance benefits of coupled multi-agent algorithms with the precomputed trajectories of the bridges to handle challenging scenarios. In practice, our approach can perform global navigation for tens to hundreds of agents on a single CPU core in 2D and 3D workspaces.

A large class of motions can be decomposed into a movement task and null-space policy subject to a set of constraints. When learning such motions from demonstrations, we aim to achieve generalisation across different unseen constraints and to increase the robustness to noise while keeping the computational cost low. There exists a variety of methods for learning the movement policy and the constraints. The effectiveness of these techniques has been demonstrated in low-dimensional scenarios and simple motions. In this paper, we present a fast and accurate approach to learning constraints from observations. This novel formulation of the problem allows the constraint learning method to be coupled with the policy learning method to improve policy learning accuracy, which enables us to learn more complex motions. We demonstrate our approach by learning a complex surface wiping policy in a 7-DOF robotic arm.

This work addresses the problem of kinematic trajectory planning for mobile manipulators with non-holonomic constraints, and holonomic operational-space tracking constraints. We obtain whole-body trajectories and time-varying kinematic feedback controllers by solving a Constrained Sequential Linear Quadratic Optimal Control problem. The employed algorithm features high efficiency through a continuous-time formulation that benefits from adaptive step-size integrators and through linear complexity in the number of integration steps. In a first application example, we solve kinematic trajectory planning problems for a 26 DoF wheeled robot. In a second example, we apply Constrained SLQ to a real-world mobile manipulator in a receding-horizon optimal control fashion, where we obtain optimal controllers and plans at rates up to 100 Hz.