ICRA 2018 Papers

Whereas various things are connected via networks thanks to IoT technology, and the era of the 4th industrial revolution which realizes optimization and efficiency by utilizing AI and big data is emerging, manufacturing systems are also significantly transforming globally. In addition to the robot revolution, along with progressing digitalization, manufacturing sites in Japan are changing, aiming at establishment of “Connected Industries” that is a solution-oriented industrial society, based on the high technological capabilities. In order to respond timely to diversifying demands of customers, it is necessary to build a next-generation manufacturing systems that realizes flexible and high productivity, such as human-robot collaboration, and it is becoming difficult to respond to the necessity by conventional safety concept. Therefore, in order to realize the 4th industrial revolution, the robot revolution, and “Connected Industries,” it is essential to establish a new safety concept corresponding to the next-generation manufacturing systems to ensure safety. This paper introduces the safety concept to be changed along with the evolution of manufacturing sites, and proposes a new safety concept, which realizes collaboration safety of humans and robots, and an outline of its safety level, for the first time in the world.

Gastric ulcer is a chronic and complex (and often complete) erosion of the stomach wall that happens as a complication of a previous chronic, inflammatory process. It represents a catastrophic situation in which the patient is critical and its conditions need to be treated fast. This study presents a remotely navigatable and deployable ingestible patch and plug for gastric ulcer treatment. The patch/plug structure is made of agarose hydrogel that can change rigidity through hydration and dehydration. When dehydrated, it is rigid and can maintain a folded configuration so it can be ingested as a “pill”. This can be guided to the targeted location by a magnetic field, and be deployed instantly by hydration, namely by supplying water from the mouth. Due to the deployable origami design, it exhibits an expansion of 10 times its initial surface area, making the device suitable for the use of dressing a surface as a patch, and filling a hole as a plug.

In this paper, we develop a method for determining the probability of a multi-robot system (MRS) being secure when robot interactions are modeled as a probabilistic graph. To define the security of an MRS, we apply an existing control-theoretic notion of network attacks based on the left invertibility of a dynamical system. We then extend previous work by assuming probabilistic robot communication and sensing, modeling the effects of noise, failure, or adversarial influence on interactions in the network. The probabilistic graph security problem can then be seen as a variant of problems solved in the field of system reliability. This interpretation motivates the application of an efficient graphical representation of boolean functions known as binary decision diagrams (BDDs). Specifically, we use the canonical properties of a special type of BDD, the Reduced Order BDD, to generate a tree that can be efficiently traversed to compute the probability that a networked MRS is left invertible, and thus, secure. We then show how our adopted approach can be applied to systems that have interactions that change over time, e.g., for mobile multi-robot teams. Finally, we demonstrate the validity of our method by simulating a mobile MRS performing a rendezvous objective, while tracking its probability of security over time.

In this paper we present ProSLAM, a lightweight open-source stereo visual SLAM system designed with simplicity in mind. This work stems from the experience gathered by the authors while teaching SLAM and aims at providing a highly modular system that can be easily implemented and understood. Rather than focusing on the well known mathematical aspects of stereo visual SLAM, we highlight the data structures and the algorithmic aspects required to realize such a system. We implemented ProSLAM using the C++ programming language in combination with a minimal set of standard libraries. The results of a thorough validation performed on several standard benchmark datasets show that ProSLAM achieves precision comparable to state-of-the-art approaches, while requiring substantially less computation.

Our human sense of touch enables us to manipulate our surroundings; therefore, complex robotic manipulation will require artificial tactile sensing. Typically tactile sensor arrays are used in robotics, implying that a straightforward way of interpreting multidimensional data is required. In this paper we present a simple visualisation approach based on applying principal component analysis (PCA) to systematically collected sets of tactile data. We apply the visualisation approach to 4 different types of tactile sensor, encompassing fingertips and vibrissal arrays. The results show that PCA can reveal structure and regularities in the tactile data, which also permits the use of simple classifiers such as k-NN to achieve good inference. Additionally, the Euclidean distance in principal component space gives a measure of sensitivity, which can aid visualisation and also be used to find regions in the tactile input space where the sensor is able to perceive with higher accuracy. We expect that these observations will generalise, and thus offer the potential for novel control methods based on touch.

Twisted string actuators (TSAs) are gaining popularity in modern engineering and robotic applications. However, in conventional actuators of this type, the twisted part of the string should not be in contact with any surfaces or objects because this may interfere with the propagation of twisting. This imposes significant constraint on potential applications of TSAs. In this paper, we investigated the feasibility of using TSAs inside conduit and demonstrated that the twists of the string can be fully propagated through the sheath and that consistent periodic behavior of the twisted string can be achieved. In addition, we investigated input-output position and force characteristics of TSAs for various deflection angles of the conduit, effect of lubrication on transmission efficiency, and compared it with conventional cable sliding transmission. We found that TSA has higher transmission efficiency than sliding due to decreased friction between the string and conduit, which is further improved by lubrication. We have managed to achieve 85 % of force transmission efficiency for the case of lubricated twisting, as opposed to the 71.74% for lubricated sliding.

This paper presents preliminary results of a handheld device to assist the insertion of intra-fascicular planar electrodes into a peripheral nervous system. The developed device consists of two units, a nerve holder and an electrode inserter. We introduce design considerations, features, and underlying mechanisms of the device. User tests and animal experiments show that users can easily and accurately adjust the insertion position and direction of nerve electrodes while manipulating the device in 3D, and that planar electrodes are successfully inserted into sciatic nerves of rats. We hope that the proposed device will help neural engineering researchers and scientists to simplify the surgical process and produce consistent experimental results.

This paper reports a preliminary study for use of a fully dynamic vehicle process model in combined underwater communication and navigation (cooperative navigation) of underwater vehicles equipped with an acoustic modem, attitude, and depth sensors, but lacking a Doppler velocity log (DVL), and a surface vehicle equipped with an acoustic modem and GPS. We report both simulation and at-sea experimental trials with the JHU Iver3 autonomous underwater vehicle (AUV). The case of underwater vehicle navigation without a DVL is of interest in several use-cases including (a) small and low-cost underwater vehicles for which DVLs may be impractical or infeasible due to their size and cost and (b) for missions in which the vehicle's altitude above the sea floor (or depth beneath overhead ice) exceeds the DVL acoustic bottom-lock range. To the best of our knowledge, all previous studies on cooperative navigation have reported use of a kinematic process model, which works well in the presence of frequent, high-accuracy velocity measurements, as is the case when the vehicle is equipped with a DVL. This preliminary study suggests that the dynamical process model may offer a significant advantage over the purely kinematic model in the absence of frequent, high-accuracy velocity measurements, as is the case when the submerged vehicle is not equipped with a DVL.

Advanced driver assistance systems allow for increasing user comfort and safety by sensing the environment and anticipating upcoming hazards. Often, this requires to accurately predict how situations will change. Recent approaches make simplifying assumptions on the predictive model of the Ego-Vehicle motion or assume prior knowledge, such as road topologies, to be available. However, in many urban areas this assumption is not satisfied. Furthermore, temporary changes (e.g. construction areas, vehicles parked on the street) are not considered by such models. Since many cars observe the environment with several different sensors, predictive models can benefit from them by considering environmental properties. In this work, we present an approach for an Ego-Vehicle path prediction from such sensor measurements of the static vehicle environment. Besides proposing a learned model for predicting the driver's multi-modal future path as a grid-based prediction, we derive an approach for extracting paths from it. In driver assistance systems both can be used to solve varying assistance tasks. The proposed approach is evaluated on real driving data and outperforms several baseline approaches.

During localization and mapping the success of point cloud registration can be compromised when there is an absence of geometric features or constraints in corridors or across doorways, or when the volumes scanned only partly overlap, due to occlusions or constrictions between subsequent observations. This work proposes a strategy to predict and prevent laser-based localization failure. Our solution relies on explicit analysis of the point cloud content prior to registration. A model predicting the risk of a failed alignment is learned by analysing the degree of spatial overlap between two input point clouds and the geometric constraints available within the region of overlap. We define a novel measure of alignability for these constraints. The method is evaluated against three real-world datasets and compared to baseline approaches. The experiments demonstrate how our approach can help improve the reliability of laser-based localization during exploration of unknown and cluttered man-made environments.

Subretinal injection is a delicate and complex microsurgery, which requires surgeons to inject the therapeutic substance in a pre-operatively defined and intra-operatively updated subretinal target area. Due to the lack of subretinal visual feedback, it is hard to sense the insertion depth during the procedure, thus affecting the results of surgical outcome and hindering the widespread use of this treatment. This paper presents a novel approach to estimate the 3D position of the needle under the retina using the information from microscope-integrated Intraoperative Optical Coherence Tomography (iOCT). We evaluated our approach on both tissue phantom and ex-vivo porcine eyes. Evaluation results show that the average error in distance measurement is 4.7 μm (maximum of 16.5 μm). We furthermore, verified the feasibility of the proposed method to track the insertion depth of needle in robot-assisted subretinal injection.

In contrast to cameras, lidars, GPS, and proprioceptive sensors, radars are affordable and efficient systems that operate well under variable weather and lighting conditions, require no external infrastructure, and detect long-range objects. In this paper, we present a reliable and accurate radar-only motion estimation algorithm for mobile autonomous systems. Using a frequency-modulated continuous-wave (FMCW) scanning radar, we first extract landmarks with an algorithm that accounts for unwanted effects in radar returns. To estimate relative motion, we then perform scan matching by greedily adding point correspondences based on unary descriptors and pairwise compatibility scores. Our radar odometry results are robust under a variety of conditions, including those under which visual odometry and GPS/INS fail.

Though a long-studied problem, motion segmentation has yet to migrate into practical applications. We argue that a vital step towards that goal lies in addressing motion segmentation for the specific setting of interest. To this end, this paper presents a new approach for image-based motion segmentation in the case of vehicles navigating inside an urban environment. We exploit two application-specific factors - the restricted camera movement and the known type of moving objects - to deal with the two major limiting factors - missing data and strong perspective effects - that affect most previous “generic” motion segmentation algorithms. By constraining the geometry and exploiting known semantic classes in the scene, we achieve much higher accuracy than previous approaches. In addition to the novel algorithm, we contribute a more realistic motion segmentation benchmark dataset for moving platforms by annotating real video sequences from the KITTI dataset. Experiments on this dataset and other synthetic data confirm the effectiveness of the proposed approach.

Popcorn kernels are a natural, edible, and inexpensive material that has the potential to rapidly expand with high force upon application of heat. Although this transition is irreversible, it carries potential for several robotic applications. Here, we examine relevant characteristics of three types of kernels including expansion ratio, transition temperature, popping force, compression strength, and biodegradability. We test the viability of popping by hot oil, hot air, microwaves, and direct contact with heated Nichrome wire. As kernels can change from regular to (larger) irregular shapes, we examine the change in inter-granular friction and propose their use as granular fluids in jamming actuators, without the need for a vacuum pump. Furthermore, as a proof-of-concept, we also demonstrate the use of popcorn-driven actuation in soft, compliant, and rigid-link grippers. Serving as a first introduction of popcorn into robotics, we hope this paper will inspire novel mechanisms for multi-functional designs.

Provides an abstract for each of the keynote presentations and may include a brief professional biography of each presenter. The complete presentations were not made available for publication as part of the conference proceedings.

One of the main strengths of self-reconfigurable modular robots (SRMR) is their ability to shape-shift and dynamically change their morphology. In the case of our SRMR system “Roombots”, these shapes can be quite arbitrary for a great variety of tasks while the major utility is envisioned to be self-reconfigurable furniture. As such, the ideas and inspirations from users quickly need to be translated into the final Roombots shape. This involves a multitude of separate processes and - most importantly - requires an intuitive user interface. Our current approach led to the development of a tangible user interface (TUI) which involves 3D-scanning of a shape formed by modeling clay and the necessary steps to prepare the digitized model to be formed by Roombots. The system is able to generate a solution in less than two minutes for our target use as demonstrated with various examples.

We present an algorithm for selecting when to communicate during online planning phases of coordinated multi-robot missions. The key idea is that a robot decides to request communication from another robot by reasoning over the predicted information value of communication messages over a sliding time-horizon, where communication messages are probability distributions over action sequences. We formulate this problem in the context of the recently proposed decentralised Monte Carlo tree search (Dec-MCTS) algorithm for online, decentralised multi-robot coordination. We propose a particle filter for predicting the information value, and a polynomial-time belief-space planning algorithm for finding the optimal communication schedules in an online and decentralised manner. We evaluate the benefit of informative communication planning for a multi-robot information gathering scenario with 8 simulated robots. Our results show reductions in channel utilisation of up to four-fifths with surprisingly little impact on coordination performance.

Motion planning is an extremely well-studied problem in the robotics community, yet existing work largely falls into one of two categories: computationally efficient but with few if any safety guarantees, or able to give stronger guarantees but at high computational cost. This work builds on a recent development called FaSTrack in which a slow offline computation provides a modular safety guarantee for a faster online planner. We introduce the notion of “meta-planning” in which a refined offline computation enables safe switching between different online planners. This provides autonomous systems with the ability to adapt motion plans to a priori unknown environments in real-time as sensor measurements detect new obstacles, and the flexibility to maneuver differently in the presence of obstacles than they would in free space, all while maintaining a strict safety guarantee. We demonstrate the meta-planning algorithm both in simulation and in hardware using a small Crazyflie 2.0 quadrotor.

Navigating in search and rescue environments is challenging, since a variety of terrains has to be considered. Hybrid driving-stepping locomotion, as provided by our robot Momaro, is a promising approach. Similar to other locomotion methods, it incorporates many degrees of freedom - offering high flexibility but making planning computationally expensive for larger environments. We propose a navigation planning method, which unifies different levels of representation in a single planner. In the vicinity of the robot, it provides plans with a fine resolution and a high robot state dimensionality. With increasing distance from the robot, plans become coarser and the robot state dimensionality decreases. We compensate this loss of information by enriching coarser representations with additional semantics. Experiments show that the proposed planner provides plans for large, challenging scenarios in feasible time.