ICRA 2019 Papers

Visual-inertial SLAM (VI-SLAM) requires a good initial estimation of the initial velocity, orientation with respect to gravity and gyroscope and accelerometer biases. In this paper we build on the initialization method proposed by Martinelli [1] and extended by Kaiser et al. [2], modifying it to be more general and efficient. We improve accuracy with several rounds of visual-inertial bundle adjustment, and robustify the method with novel observability and consensus tests, that discard erroneous solutions. Our results on the EuRoC dataset show that, while the original method produces scale errors up to 156%, our method is able to consistently initialize in less than two seconds with scale errors around 5%, which can be further reduced to less than 1% performing visual-inertial bundle adjustment after ten seconds.

Due to recent deep learning techniques, person detection seems to be solved in the computer vision domain, however, it is still an issue in mobile robotics. On a robot only limited computing capacities are available. The challenge gets even more difficult when operating in an environment, with people in poses different from the standard upright ones. In this work the environment of a supermarket is considered. Unlike most scenarios targeted by the community, persons not only occur in standing postures, but also grasping into the shelves or squatting in front of them. Furthermore, people are heavily occluded, e.g. by shopping carts. In such a challenging environment, it is important to perceive people early enough and in real-time in order to enable a socially aware navigation. Classical person detectors often suffer from a high posture variance or do not achieve acceptable real-time detection rates. For this reason, different components from the 3D object detection domain have been used to create a new robust person detector for mobile application. Operating on 3D point clouds allows fast detections in real-time up to our goal distance of ten meters and above using the Kinect2 depth sensor. The detector can even differentiate between typical postures of customers who stand or squat in front of shelves.

Fast Graspability Evaluation (FGE) has been proposed as a method for detecting grasping positions on objects and is now being used for industrial robots. FGE uses convolution of hand templates with regions on the target object to estimate the optimum grasping posture. However, the hand opening width and rotation angles must be set with high resolution to achieve highly accurate results and the computational load is high. To address that issue, we propose a method in which hand templates are represented in compact form for faster processing by using singular value decomposition. Applying singular value decomposition enables hand templates to be represented as linear combinations of a small number of eigenvalue templates and eigenfunctions. Eigenfunctions take discrete values, but response values can be calculated with arbitrary parameters by fitting a continuous function. Experimental results show that the proposed method reduces computation time by two thirds while maintaining the same detection accuracy as conventional FGE for both parallel hands and three-finger hands.

This paper aims to design quadrotor swarm performances, where the swarm acts as an integrated, coordinated unit embodying moving and deforming objects. We divide the task of creating a choreography into three basic steps: designing swarm motion primitives, transitioning between those movements, and synchronizing the motion of the drones. The result is a flexible framework for designing choreographies comprised of a wide variety of motions. The motion primitives can be intuitively designed using a few parameters, providing a rich library for choreography design. Moreover, we combine and adapt existing goal assignment and trajectory generation algorithms to maximize the smoothness of the transitions between motion primitives. Finally, we propose a correction algorithm to compensate for motion delays and synchronize the motion of the drones to a desired periodic motion pattern. The proposed methodology was validated experimentally by generating and executing choreographies on a swarm of 25 quadrotors.

This paper proposes a fast terminal sliding mode super twisting controller (FTSMSTC) design for quadrotor position and altitude tracking in the presence of bounded disturbances. A nonlinear fast terminal sliding manifold has been proposed for fast convergence of the tracking error to zero in finite time unlike the conventional sliding mode control (CSMC) that guarantee only asymptotic convergence of the tracking error. The super twisting reaching law has been proposed to deal with the chattering phenomena, which is inherent in the CSMC. The finite time stability of the complete closed loop system is investigated using Lyapunov stability theory and an analytical expression for the convergence time has also been derived. The effectiveness of the designed controller is checked against the CSMC using MATLAB simulation. The controller has been experimentally validated using the DJI Matrice M100 as a proof of utility in real time applications.

Path planners are generally categorised as either trajectory optimisers or sampling-based planners. The latter is the predominant planning paradigm for occupancy maps. Most trajectory optimisers require a fully defined artificial potential field for planning and cannot incorporate updates from a partially observed model such as an occupancy map. A stochastic trajectory optimiser capable of planning over occupancy map was presented in [1]. However, its scalability is limited by the cubic complexity of the Gaussian process path representation. In this work, we introduce a novel highly expressive path representation based on kernel approximation to perform trajectory optimisation over occupancy maps. This approach reduces the computational complexity to a fixed cost that only depends on the number of features. We show that stochastic sampling is crucial for planning in occupancy maps and present comparisons to other state-of-the-art planning methods, using simulated and real occupancy data. These experiments demonstrate the significant reduction in runtime, resulting in performance comparable to or better than sampling-based methods.

This paper is about fast motion estimation with scanning radar. We use weak supervision to train a focus of attention policy which actively down-samples the measurement stream before data association steps are undertaken. At training, we avoid laborious manual labelling by exploiting short-term sensor coherence from multiple poses in the presence of an external ego-motion estimator (for example, wheel odometry). In this way, we generate copious annotated measurements which can be used for training a learning algorithm in a weakly-supervised fashion. We demonstrate the validity of the approach in the context of a Radar Odometry (RO) task, pre-filtering raw data with a popular image segmentation network trained as presented. We evaluate our system against 26 km of data collected in Central Oxford and show consistent motion estimation with greatly reduced radar processing times (by a factor of 2.36).

Augmenting a robot with the capacity to understand the activities of the people it collaborates with in order to then label and segment those activities allows the robot to generate an efficient and safe plan for performing its own actions. In this work, we introduce an online activity segmentation algorithm that can detect activity segments by processing a partial trajectory. We model the transitions through activities as a hidden Markov model, which runs online by implementing an efficient particle-filtering approach to infer the maximum a posteriori estimate of the activity sequence. This process is complemented by an online search process to refine activity segments using task model information about the partial order of activities. We evaluated our algorithm by comparing its performance to two state-of-the-art activity segmentation algorithms on three human activity datasets. The proposed algorithm improved activity segmentation accuracy across all three datasets compared with the other two approaches, with a range from 11.3% to 65.5%, and could accurately recognize an activity through observation alone for 31.6% of the initial trajectory of that activity, on average. We also implemented the algorithm onto an industrial mobile robot during an automotive assembly task in which the robot tracked a human worker's progress and provided the worker with the correct materials at the appropriate time.

We present an efficient algorithm for motion planning and controlling a robot system with a high number of degrees-of-freedom (DOF). These systems include high-DOF soft robots and articulated robots interacting with a deformable environment. We present a novel technique to accelerate the evaluations of the forward dynamics function by storing the results of costly computations in a hierarchical adaptive grid. Furthermore, we exploit the underactuated properties of the robot systems and build the grid in a low-dimensional space. Our approach approximates the forward dynamics function with guaranteed error bounds and can be used in optimization-based motion planning and reinforcement-learning-based feed-back control. We highlight the performance on two high-DOF robot systems: a line-actuated elastic robot arm and an underwater swimming robot in water. Compared to prior techniques based on exact dynamics evaluation, we observe one to two orders of magnitude improvement in the performance.

Semantic scene understanding is important for autonomous robots that aim to navigate dynamic environments, manipulate objects, or interact with humans in a natural way. In this paper, we address the problem of jointly performing semantic segmentation as well as instance segmentation in an online fashion, so that autonomous robots can use this information on-the-go and without sacrificing accuracy. We achieve this by exploiting a local connectivity prior of objects in the real world and a multi-task convolutional neural network architecture. The network identifies the individual object instances and their classes without region proposals or pre-segmentation of the images into individual classes. We implemented and thoroughly evaluated our approach, and our experiments suggest that our method can be used to accurately segment instance masks of objects and identify their class in an online fashion.

Robots working in human environments often encounter a wide range of articulated objects, such as tools, cabinets, and other jointed objects. Such articulated objects can take an infinite number of possible poses, as a point in a potentially high-dimensional continuous space. A robot must perceive this continuous pose in order to manipulate the object to a desired pose. This problem of perception and manipulation of articulated objects remains a challenge due to its high dimensionality and multi-modal uncertainty. In this paper, we propose a factored approach to estimate the poses of articulated objects using an efficient non-parametric belief propagation algorithm. We consider inputs as geometrical models with articulation constraints, and observed 3D sensor data. The proposed framework produces object-part pose beliefs iteratively. The problem is formulated as a pairwise Markov Random Field (MRF) where each hidden node (continuous pose variable) models an observed object-part's pose and each edge denotes an articulation constraint between a pair of parts. We propose articulated pose estimation by a Pull Message Passing algorithm for Nonparametric Belief Propagation (PMPNBP) and evaluate its convergence properties over scenes with articulated objects.

Scarce data is a major challenge to scaling robot learning to truly complex tasks, as we need to generalize locally learned policies over different task contexts. Contextual policy search offers data-efficient learning and generalization by explicitly conditioning the policy on a parametric context space. In this paper, we further structure the contextual policy representation. We propose to factor contexts into two components: target contexts that describe the task objectives, e.g. target position for throwing a ball; and environment contexts that characterize the environment, e.g. initial position or mass of the ball. Our key observation is that experience can be directly generalized over target contexts. We show that this can be easily exploited in contextual policy search algorithms. In particular, we apply factorization to a Bayesian optimization approach to contextual policy search both in sampling-based and active learning settings. Our simulation results show faster learning and better generalization in various robotic domains. See our supplementary video: https://youtu.be/IIJTbBAOufDY.

Bio-syncretic robots made up of living biological systems and electromechanical systems may have the potential excellent performance of natural biological entities. Therefore, the study of the bio-syncretic robots has got lots of attention in recent years. The 3D skeletal muscles have been used widely, due to the considerable contraction force and the controllability. However, the low differentiation quality of the C2C12 in the tissues hinders the broad application in the development of the skeleton muscle actuated bio-syncretic robots. In this work, an approach based on circular mould and rotary electrical stimulation to build high-quality muscle rings, which can be used to actuate various bio-syncretic robots, has been proposed. Firstly, the advantage of the proposed circular mould for the muscle rings culture has been shown by simulation. Then, the muscle rings have been fabricated with different moulds using the experiment-optimized compositions of the biological mixture. After that, the muscle rings in the circular moulds with different electrical stimulations have been cultured, to show the superiority of the proposed rotary electrical stimulation. Moreover, the contractility of the muscle rings have been measured under the different electrical pulses stimulation, for the study of the control property of the muscle rings. This work may be meaningful not only the development of bio-syncretic robots actuated by 3D muscle tissues but also the muscle tissue engineering.

This paper presents the design and development of a highly articulated, continuum, wearable, fabric-based Soft Poly-Limb (fSPL). This fabric soft arm acts as an additional limb that provides the wearer with mobile manipulation assistance through the use of soft actuators made with high-strength inflatable fabrics. In this work, a set of systematic design rules is presented for the creation of highly compliant soft robotic limbs through an understanding of the fabric based components behavior as a function of input pressure. These design rules are generated by investigating a range of parameters through computational finite-element method (FEM) models focusing on the fSPL's articulation capabilities and payload capacity in 3D space. The theoretical motion and payload outputs of the fSPL and its components are experimentally validated as well as additional evaluations verify its capability to safely carry loads 10.1x its body weight, by wrapping around the object. Finally, we demonstrate how the fully collapsible fSPL can comfortably be stored in a soft-waist belt and interact with the wearer through spatial mobility and preliminary pick-and-place control experiments.

Information-based mapping algorithms are critical to robot exploration tasks in several applications ranging from disaster response to space exploration. Unfortunately, most existing information-based mapping algorithms are plagued by the computational difficulty of evaluating the Shannon mutual information between potential future sensor measurements and the map. This has lead researchers to develop approximate methods, such as Cauchy-Schwarz Quadratic Mutual Information (CSQMI). In this paper, we propose a new algorithm, called Fast Shannon Mutual Information (FSMI), which is significantly faster than existing methods at computing the exact Shannon mutual information. The key insight behind FSMI is recognizing that the integral over the sensor beam can be evaluated analytically, removing an expensive numerical integration. In addition, we provide a number of approximation techniques for FSMI, which significantly improve computation time. Equipped with these approximation techniques, the FSMI algorithm is more than three orders of magnitude faster than the existing computation for Shannon mutual information; it also outperforms the CSQMI algorithm significantly, being roughly twice as fast, in our experiments.

We propose a novel stereo visual SLAM framework considering both accuracy and speed at the same time. The framework makes full use of the advantages of key-feature-based multiple view geometry (MVG) and direct-based formulation. At the front-end, our system performs direct formulation and constant motion model to predict a robust initial pose, reprojects local map to find 3D-2D correspondence and finally refines pose by the reprojection error minimization. This frontend process makes our system faster. At the back-end, MVG is used to estimate 3D structure. When a new keyframe is inserted, new mappoints are generated by triangulating. In order to improve the accuracy of the proposed system, bad mappoints are removed and a global map is kept by bundle adjustment. Especially, the stereo constraint is performed to optimize the map. This back-end process makes our system more accurate. Experimental evaluation on EuRoC dataset shows that the proposed algorithm can run at more than 100 frames per second on a consumer computer while achieving highly competitive accuracy.

External wrench estimation is very helpful for aerial exploration and manipulation tasks. During the exploration, there might be unseen obstacles to cause dangerous collisions. The estimation of the external force and torque is also beneficial in aerial manipulation tasks. In this paper, we present a framework of estimating the external wrench for the aerial multilink robot based on the onboard inertial measurement unit (IMU) sensors, joints state and robot dynamic models. Compared to the conventional multirotor robot, the center of mass (CoM) is always changing when the robot transforms. The sensor could not be attached to CoM to observe the acceleration data. Consequently, we present a novel method by applying a distributed IMU system to estimate the CoM linear and angular accelerations for the external wrench estimation. With the help of the robot model, the position of the contact point could be estimated, which is useful in exploring tasks to safely interact with the physical world. We design the contact-aided navigation strategy and computationally efficient motion primitives library to help our robot react to the unexpected collision. We experimentally validate our framework with a two-dimensional multilink aerial robot to show the results of external wrench estimator and its further applications2.2Experiment video: https://youtu.be/R-WDReLnWWI

We propose a highly flexible approach to human-robot cooperation, where a robot dynamically selects operations contributing to a shared goal from a given task model. Therefore, knowledge on the task progress is extracted from a world model constructed from eye-in-hand camera images. Data generated from such partial workspace observations is not reliable over time, as humans may interact with resources. We therefore use a human-aware world model maintaining a measure for trust in stored objects regarding recent human presence and previous task progress. Our contribution is an action selection algorithm that uses this trust measure to interleave task operations with active vision to refresh the world model. Large-scale experiments cover various sorts of human participation in different benchmark tasks through simulation of simplified, partially randomized human models. Results illuminate system behaviour and performance for different parametrizations of our human-robot teaming framework.

Service robots require the ability to recognize various household objects in order to carry out certain tasks, such as fetching an object for a person. Manually collecting information on all the objects a robot may encounter in a household is tedious and time-consuming; therefore this paper proposes the use of large-scale data from existing trademark databases. These databases contain logo images and a description of the goods and services the logo was registered under. For example, Pepsi is registered under soft drinks. We extend domain randomization in order to generate synthetic data to train a convolutional neural network logo detector, which outperformed previous logo detectors trained on synthetic data. We also provide a practical implementation for object fetching on a robot, which uses a Kinect and the logo detector to identify the object the human user requested. Tests on this robot indicate promising results, despite not using any real world photos for training.