ICRA 2018 Papers

In this work, we present an extension to a linear Model Predictive Control (MPC) scheme that plans external contact forces for the robot when given multiple contact locations and their corresponding friction cone. To this end, we set up a two-step optimization problem. In the first optimization, we compute the Center of Mass (CoM) trajectory, foot step locations, and introduce slack variables to account for violating the imposed constraints on the Zero Moment Point (ZMP). We then use the slack variables to trigger the second optimization, in which we calculate the optimal external force that compensates for the ZMP tracking error. This optimization considers multiple contacts positions within the environment by formulating the problem as a Mixed Integer Quadratic Program (MIQP) that can be solved at a speed between 100-300 Hz. Once contact is created, the MIQP reduces to a single Quadratic Program (QP) that can be solved in real-time (<; 1kHz). Simulations show that the presented walking control scheme can withstand disturbances 2-3× larger with the additional force provided by a hand contact.

The control problem in self-adjustable treadmills is to keep the position of the user at a reference position. The position error is the key obstacle in facilitating the abrupt stopping on self-adjustable treadmills. Another difficulty in this application is the oscillatory response of the closed-loop system. The conventional control methods utilize a high-gain observer to estimate the user velocity and exploit this information beside feedback signals to decrease the position error. Utilizing the high-gain observer, however, applies anomalous force (AF) to the user, leading to an unnatural feeling, and does not guarantee an oscillation-free response for the output. This paper aims to alleviate these problems by proposing a supervisory control scheme. First, a RISE controller is utilized for walking/running stage to compensate for slowly varying uncertainties in the system model without applying a large AF. Then, a positive-output controller is exploited for the stopping stage to guarantee the convergence of the position error without oscillation. Using the estimated intentional velocity and acceleration, a supervisory system is designed to switch between the controllers. Experimental results show the superiority of the proposed approach over the existing methods.

Solving the direct kinematics (DK) of cable-driven parallel robots (CDPR) based only on the cable length measurements is a demanding problem that is still not well mastered, especially for robots having sagging cables. A model-based approach may be used to solve this problem but the model parameters and measurements are uncertain, thereby leading to positioning inaccuracy. A possible way to improve the accuracy and speed up the solving is to add extra measurements. For that purpose a preliminary step is to determine what type of measurements are possible and then to estimate how accurate they are. For that purpose we have used a CDPR with 4 cables that has been instrumented with various types of extra measurements: cable tensions and orientations, platform orientation. Ground truth has been established and we have compared the data provided by the extra sensors with their real values. This work shows that cable tensions sensors and platform orientation sensors are not good candidates to be used for the DK while cable orientations may be obtained with a good accuracy both in static poses or during a quasi-static motion.

In this paper we address the problem of controlling a robotic system mounted on an actuated floating base for space applications. In particular, we investigate the stability issues due to the low rate of the base control unit. We propose a passivity-based stabilizing controller based on the time domain passivity approach. The controller uses a variable damper regulated by a designed energy observer. The effectiveness of the proposed strategy is validated on a base-manipulator multibody simulation.

Ground effect refers to the apparent increase in lift that an aircraft experiences when it flies close to the ground. For helicopters, this effect has been modeled since the 1950's based on the work of Cheeseman and Bennett, perhaps the most common method for predicting hover performance due to ground effect. This model, however, is based on assumptions that do not hold for small-scale rotorcraft because it was developed specifically for conventional helicopters. It is not clear if the Cheeseman-Bennett model can be applied to today's multirotor UAVs. In this paper, we compare the Cheeseman-Bennett model to experimental results for rotor performance due to ground effect in several small-scale multirotor and single-rotor configurations. Experimental findings suggest that some of the conventional thinking surrounding helicopter ground effect cannot be applied directly to rotorcraft using fixed propellers at variable speeds (e.g. multirotors), and that it is necessary to adjust the helicopter models to better reflect the differences in such aircraft. The experimental results for multirotors presented are for multiple propeller configurations, speeds and spacings. Ultimately, this work will facilitate the development of an improved UAV flight controller that can accurately account for ground effect to improve flight stability near surfaces and structures.

Mesh plays an indispensable role in dense realtime reconstruction essential in robotics. Efforts have been made to maintain flexible data structures for 3D data fusion, yet an efficient incremental framework specifically designed for online mesh storage and manipulation is missing. We propose a novel framework to compactly generate, update, and refine mesh for scene reconstruction upon a volumetric representation. Maintaining a spatial-hashed field of cubes, we distribute vertices with continuous value on discrete edges that support O(1) vertex accessing and forbid memory redundancy. By introducing Hamming distance in mesh refinement, we further improve the mesh quality regarding the triangle type consistency with a low cost. Lock-based and lock-free operations were applied to avoid thread conflicts in GPU parallel computation. Experiments demonstrate that the mesh memory consumption is significantly reduced while the running speed is kept in the online reconstruction process.

In this paper, we are concerned with the automation of single robot exogeneous assembly in simple planar settings. Here, a set of of unactuated disk-shaped parts needs to be serviced by a single robot that resides outside the workspace and is capable of moving the parts one at a time through sliding or lifting slightly. The task is to have the parts end up in their respective goal positions without any collisions whilst moving. We present an automated reactive approach through the composition of one-part movements. The movement of each part is achieved by a controller obtained through the projection of a carefully constructed vector field on the respective configuration subspace. Such a scheme is known to accommodate positional variations naturally. Once the movement terminates, the robot chooses the next part to move in a cyclic manner. The contribution of this paper is to show for the first time that with certain restrictions on the allowed goal configurations, the assembly task is either successfully completed or terminated (rather than useless cycling of a part or from part to part) while the generated sequence of motions never causes collisions among the parts.

We study the problem of finding a tour for an energy-limited Unmanned Aerial Vehicle (UAV) to visit a set of sites in the least amount of time. We envision scenarios where the UAV can be recharged along the way either by landing on stationary recharging stations or on Unmanned Ground Vehicles (UGVs) acting as mobile recharging stations. This leads to a new variant of the Traveling Salesperson Problem (TSP). We present an algorithm that finds not only the order in which to visit the sites but also when and where to land on the charging stations to recharge. Our algorithm plans tours for the UGVs as well as determines best locations to place stationary charging stations. While the problems we study are NP-Hard, we present a practical solution using Generalized TSP that finds the optimal solution. If the UGVs are slower, the algorithm also finds the minimum number of UGVs required to support the UAV mission such that the UAV is not required to wait for the UGV. Our simulation results show that the running time is acceptable for reasonably sized instances.

Motivated by applications in multirobot team selection, in this paper, we present a novel algorithm for computing optimal solution of chance-constrained 0-1 knapsack problem. In this variation of the knapsack problem, the objective function is deterministic but the weights of the items are stochastic and therefore the knapsack constraint is stochastic. We convert the chance-constrained knapsack problem to a two-dimensional discrete optimization problem on the variance-mean plane, where each point on the plane can be identified with an assignment of items to the knapsack. By exploiting the geometry of the non-convex feasible region of the chance-constrained knapsack problem in the variance-mean plane, we present a novel deterministic technique to find an optimal solution by solving a sequence of deterministic knapsack problems (called risk-averse knapsack problem). We apply our algorithm to a multirobot team selection problem to cover a given route, where the length of the route is much larger than the length each individual robot can fly and the length that an individual robot can fly is a random variable (with known mean and variance). We present simulation results on randomly generated data to demonstrate that our approach is scalable with both the number of robots and increasing uncertainty of the distance an individual robot can travel.

In this paper, we perform observability analysis for inertial navigation systems (INS) aided by generic exteroceptive range and/or bearing sensors with different geometric features including points, lines and planes. While the observability of vision-aided INS (VINS, which uses camera as a bearing sensor) with point features has been extensively studied in the literature, we analytically show that the same observability property remains if using generic range and/or bearing measurements, and if global measurements are also available, as expected, some unobservable directions dismiss. We study in-depth the effects of four degenerate motions on the system observability. In particular, building upon the observability analysis of the aided INS with point features, we perform observability analysis for the same system but with line and plane features, respectively, and show that there exist 5 (and 6) unobservable directions for a single line (and plane) feature. Moreover, we, for the first time, analytically derive the unobservable directions for the cases of multiple lines/planes. We validate our analysis through Monte Carlo simulations.

This paper presents a controller for the vertical height of a hopping robot. The ability to accurately change the hop height every step will contribute toward the traversal of discontinuous terrain with limited safe footholds, with application to bipedal or quadrupedal running. A key feature of the approach presented is the use of information from previous hops/steps to inform the control of the current step. As well as avoiding modelling errors, this allows the robot to make on-line adjustments in response to changes in system parameters or the environment. The algorithm is simple enough to be easily implemented on a low power hardware, not requiring computationally demanding optimisation or numerical simulation. The effectiveness of this approach has been demonstrated for constrained vertical hopping in simulation and on a pneumatically actuated hopper.

We propose AffordanceNet, a new deep learning approach to simultaneously detect multiple objects and their affordances from RGB images. Our AffordanceNet has two branches: an object detection branch to localize and classify the object, and an affordance detection branch to assign each pixel in the object to its most probable affordance label. The proposed framework employs three key components for effectively handling the multiclass problem in the affordance mask: a sequence of deconvolutional layers, a robust resizing strategy, and a multi-task loss function. The experimental results on the public datasets show that our AffordanceNet outperforms recent state-of-the-art methods by a fair margin, while its end-to-end architecture allows the inference at the speed of 150ms per image. This makes our AffordanceNet well suitable for real-time robotic applications. Furthermore, we demonstrate the effectiveness of AffordanceNet in different testing environments and in real robotic applications. The source code is available at https://github.com/nqanh/affordance-net.

Despite impressive advances of humanoid robotics, the autonomous planning of whole-body loco-manipulation actions in unknown environments is still an open problem. In our previous work, we addressed two fundamental aspects related to this problem: 1) the autonomous detection of end-effector contact opportunities in unknown environments and 2) the goal-directed planning of multi-contact pose sequences, which can serve as the starting point for motion planning and control approaches of reduced complexity. Both problems suffer from the extensive amounts of possible solutions, particularly due to the complexity of humanoid robots and the multitude of available contact opportunities. In this paper, we propose a method for the planning of whole-body multi-contact tasks based on our previous work on vision-based detection of loco-manipulation affordances and whole-body multi-contact pose sequence planning. We demonstrate a combined approach for planning multi-contact pose sequences with a focus on the utilization of available end-effectors for stabilizing contacts with the environment during loco-manipulation tasks. The method is evaluated in simulation in multiple exemplary scenarios based on actual sensor data and the humanoid robot ARMAR-4.

In this paper, we present the achievement of aerial grasping by shape adaptive transformation to the object shape, using a novel transformable aerial robot called HALO: Horizontal Plane Transformable Aerial Robot with Closed-loop Multilinks Structure. Aerial manipulation is an active research area and using multiple aerial robots is an effective solution for the large size object. However the cooperation is considered that there are some difficulties such as the synchronized flight control and collision with each other. Then, we focus on the transformable aerial robot with two-dimensional multilinks proposed in our previous works, which can transform to the suitable form for the target object and grasp it. However the transformable aerial robot with the serial-link structure could not achieve stable flight in terms of horizontal position and yaw control due to the low rigidity and large inertia in the case of more than 4 links. Thus, first we construct a novel type of multilinks with closed-loop structure to avoid the deformation and a new link module with a tilted propeller for fully-actuated control. Second, we describe transformation method with closed-loop multilinks. Third, we present the optimization planning method for the multilinks form to be adaptive to the two-dimensional shape of the target object. Finally, we present experimental results to demonstrate the feasibility of closed-loop aerial transformation and aerial grasping for the large size object.

We present a method of improving visual place recognition and metric localisation under very strong appearance change. We learn an invertable generator that can transform the conditions of images, e.g. from day to night, summer to winter etc. This image transforming filter is explicitly designed to aid and abet feature-matching using a new loss based on SURF detector and dense descriptor maps. A network is trained to output synthetic images optimised for feature matching given only an input RGB image, and these generated images are used to localize the robot against a previously built map using traditional sparse matching approaches. We benchmark our results using multiple traversals of the Oxford RobotCar Dataset over a year-long period, using one traversal as a map and the other to localise. We show that this method significantly improves place recognition and localisation under changing and adverse conditions, while reducing the number of mapping runs needed to successfully achieve reliable localisation.

Place recognition is an essential component of Simultaneous Localization And Mapping (SLAM). Under severe appearance change, reliable place recognition is a difficult perception task since the same place is perceptually very different in the morning, at night, or over different seasons. This work addresses place recognition as a domain translation task. Using a pair of coupled Generative Adversarial Networks (GANs), we show that it is possible to generate the appearance of one domain (such as summer) from another (such as winter) without requiring image-to-image correspondences across the domains. Mapping between domains is learned from sets of images in each domain without knowing the instance-to-instance correspondence by enforcing a cyclic consistency constraint. In the process, meaningful feature spaces are learned for each domain, the distances in which can be used for the task of place recognition. Experiments show that learned features correspond to visual similarity and can be effectively used for place recognition across seasons.

We consider the problem of online environmental sampling and modeling for multi-robot sensor coverage, where a team of robots spread out over the workspace in order to optimize the overall sensing performance. In contrast to most existing works on multi-robot coverage control that assume prior knowledge of the distribution of environmental phenomenon, also known as density function, we relax this assumption and enable the robot team to efficiently learn the model of the unknown density function Online using adaptive sampling and non-parametric inference such as Gaussian Process (GP). To capture significantly different components of the environmental phenomenon, we propose a new approach with mixture of locally learned Gaussian Processes for collective model learning and an information-theoretic criterion for simultaneous adaptive sampling in multi-robot coverage. Our approach demonstrates a better generalization of the environment modeling and thus the improved performance of coverage without assuming the density function is known a priori. We demonstrate the effectiveness of our algorithm via simulations of information gathering from indoor static sensors.

We developed a robotic lower-limb exoskeleton for those who have weakened muscle due to aging and experience difficulty in walking or getting up without help. The exoskeleton covering both limbs from the feet to the waist has 6 electric actuators in the hip abduction/adduction, hip extension/flexion and knee extension/flexion joints. For users with volitional motion, delivering assistance power according to their intention is a challenging task. We propose an adaptive oscillator-based controller to assist users walk in the lower-limb exoskeleton. To adapt to changes in walking speed and environment, motion command from the controller is modulated by estimate walking speed and walking environment recognized as one of the following categories: level ground, stairs up/down and slope up/down. Experimental results demonstrate the feasibility of the proposed environment recognition method and the impact of assistance on the metabolic cost of walking on level and inclined treadmills.

A commercial robot, trained by its manufacturer to recognize a predefined number and type of objects, might be used in many settings, that will in general differ in their illumination conditions, background, type and degree of clutter, and so on. Recent computer vision works tackle this generalization issue through domain adaptation methods, assuming as source the visual domain where the system is trained and as target the domain of deployment. All approaches assume to have access to images from all classes of the target during training, an unrealistic condition in robotics applications. We address this issue proposing an algorithm that takes into account the specific needs of robot vision. Our intuition is that the nature of the domain shift experienced mostly in robotics is local. We exploit this through the learning of maps that spatially ground the domain and quantify the degree of shift, embedded into an end-to-end deep domain adaptation architecture. By explicitly localizing the roots of the domain shift we significantly reduce the number of parameters of the architecture to tune, we gain the flexibility necessary to deal with subset of categories in the target domain at training time, and we provide a clear feedback on the rationale behind any classification decision, which can be exploited in human-robot interactions. Experiments on two different settings of the iCub World database confirm the suitability of our method for robot vision.