ICRA 2019 Papers

This paper presents a new method to efficiently monitor localization safety in mobile robots. Localization safety is quantified by measuring the system's integrity risk, which is a well-known aviation performance metric. However, aviation integrity monitoring solutions almost exclusively rely on the Global Navigation Satellite System (GNSS) while robot navigation usually needs the additional information provided by a state evolution model and/or relative positioning sensors, which makes previously established approaches impractical. In response, this paper develops an efficient integrity monitoring methodology applicable to Kalman Filter-based localization. The work is intended for life-or mission-critical operations such as co-robot applications where ignoring the impact of faults can jeopardize human safety.

Humanoid robots dynamically navigate an environment by interacting with it via contact wrenches exerted at intermittent contact poses. Therefore, it is important to consider dynamics when planning a contact sequence. Traditional contact planning approaches assume a quasi-static balance criterion to reduce the computational challenges of selecting a contact sequence over a rough terrain. This however limits the applicability of the approach when dynamic motions are required, such as when walking down a steep slope or crossing a wide gap. Recent methods overcome this limitation with the help of efficient mixed integer convex programming solvers capable of synthesizing dynamic contact sequences. Nevertheless, its exponential-time complexity limits its applicability to short time horizon contact sequences within small environments. In this paper, we go beyond current approaches by learning a prediction of the dynamic evolution of the robot centroidal momenta, which can then be used for quickly generating dynamically robust contact sequences for robots with arms and legs using a search-based contact planner. We demonstrate the efficiency and quality of the results of the proposed approach in a set of dynamically challenging scenarios.

We present an algorithm for combining natural language processing (NLP) and fast robot motion planning to automatically generate robot movements. Our formulation uses a novel concept called Dynamic Constraint Mapping to transform complex, attribute-based natural language instructions into appropriate cost functions and parametric constraints for optimization-based motion planning. We generate a factor graph from natural language instructions called the Dynamic Grounding Graph (DGG), which takes latent parameters into account. The coefficients of this factor graph are learned based on conditional random fields (CRFs) and are used to dynamically generate the constraints for motion planning. We map the cost function directly to the motion parameters of the planner and compute smooth trajectories in dynamic scenes. We highlight the performance of our approach in a simulated environment and via a human interacting with a 7-DOF Fetch robot using intricate language commands including negation, orientation specification, and distance constraints.

Collision detection has attracted attention of researchers for decades in the field of computer graphics, robot motion planning, computer aided design, etc. A large number of successful algorithms have been proposed and applied, which make use of convex polytopes and bounding volumes as primitives. However, algorithms for those shapes rely significantly on the complexity of the meshes. This paper deals with collision detection for shapes with simple and exact mathematical descriptions, such as ellipsoids and superquadrics. These primitives have a wide range of applications in representing complex objects and have much fewer parameters than meshes. The foundation of the proposed collision detection scheme relies on the closed-form Minkowski sums between ellipsoids and superquadrics in n-dimensional Euclidean space. The basic idea here is to shrink the ellipsoid into a point and expand each superquadric into a new offset surface with closed-form parametric expression. The solutions for detecting relative positions between a point and a general convex differentiable parametric surface in both 2D and 3D are derived, leading to an algorithm for exact collision detection. To compare between exact and inexact algorithms, an accuracy metric is introduced based on the Principal Kinematic Formula (PKF). The proposed algorithm is then compared with existing wellknown algorithms: Gilbert-Johnson-Keerthi (GJK) and Algebraic Separation Conditions (ASC). The results show that the proposed algorithm performs competitively with these efficient checkers.

Data association in SLAM is fundamentally challenging, and handling ambiguity well is crucial to achieve robust operation in real-world environments. When ambiguous measurements arise, conservatism often mandates that the measurement is discarded or a new landmark is initialized rather than risking an incorrect association. To address the inevitable “duplicate” landmarks that arise, we present an efficient map-merging framework to detect duplicate constellations of landmarks, providing a high-confidence loop-closure mechanism well-suited for object-level SLAM. This approach uses an incrementally-computable approximation of landmark uncertainty that only depends on local information in the SLAM graph, avoiding expensive recovery of the full system covariance matrix. This enables a search based on geometric consistency (GC) (rather than full joint compatibility (JC)) that inexpensively reduces the search space to a handful of “best” hypotheses. Furthermore, we reformulate the commonly-used interpretation tree to allow for more efficient integration of clique-based pairwise compatibility, accelerating the branch-and-bound max-cardinality search. Our method is demonstrated to match the performance of full JC methods at significantly-reduced computational cost, facilitating robust object-based loop-closure over large SLAM problems.

A method is presented for solving the discrete-time finite-horizon Linear Quadratic Regulator (LQR) problem subject to auxiliary linear equality constraints, such as fixed end-point constraints. The method explicitly determines an affine relationship between the control and state variables, as in standard Riccati recursion, giving rise to feedback control policies that account for constraints. Since the linearly-constrained LQR problem arises commonly in robotic trajectory optimization, having a method that can efficiently compute these solutions is important. We demonstrate some of the useful properties and interpretations of said control policies, and we compare the computation time and complexity of our method against existing methods.

Visual localization, i.e., determining the position and orientation of a vehicle with respect to a map, is a key problem in autonomous driving. We present a multi-camera visual inertial localization algorithm for large scale environments. To efficiently and effectively match features against a pre-built global 3D map, we propose a prioritized feature matching scheme for multi-camera systems. In contrast to existing works, designed for monocular cameras, we (1) tailor the prioritization function to the multi-camera setup and (2) run feature matching and pose estimation in parallel. This significantly accelerates the matching and pose estimation stages and allows us to dynamically adapt the matching efforts based on the surrounding environment. In addition, we show how pose priors can be integrated into the localization system to increase efficiency and robustness. Finally, we extend our algorithm by fusing the absolute pose estimates with motion estimates from a multi-camera visual inertial odometry pipeline (VIO). This results in a system that provides reliable and drift-less pose estimation. Extensive experiments show that our localization runs fast and robust under varying conditions, and that our extended algorithm enables reliable real-time pose estimation.

In this paper, we investigated how the stiffness and damping properties of soft robotic feet affect the stability and energetic economy of bipedal robotic walking. To this end, we manufactured four different spherical feet from the following materials: hollow rubber, Sorbothane, Norsorex, and Neoprene. The materials were specifically chosen to cover a wide range of stiffness and damping values. The impact response of each design was first characterized in a drop test rig. We then evaluated the performance of each foot in an extensive series of walking experiments on the planar bipedal robot RAM one. Our results showed that, at low speeds, the feet with lower damping had a smaller energy cost of walking, possibly due to greater return of mechanical energy at lift-off. However, at speeds above 0.5m\s, the feet with lower damping started to exhibit a bouncing behaviour which led to higher walking instability and increased the energy cost of walking. Additionally, we found the feet with lower stiffness to be more economical across all walking speeds. Our results provide insight into the role of foot properties in bipedal walking and may help with the design of walking robots.

Physical Human-Robot Collaboration (pHRC) is about the interaction between one or more human operator(s) and one or more robot(s) in direct contact and voluntarily exchanging forces to accomplish a common task. In any pHRC, the intuitiveness of the interaction has always been a priority, so that the operator can comfortably and safely interact with the robot. So far, the intuitiveness has always been described in a qualitative way. In this paper, we suggest an objective way to evaluate intuitiveness, known as prediction error negativity (PEN) using electroencephalogram (EEG). PEN is defined as a negative deflection in event related potential (ERP) due to cognitive conflict, as a consequence of a mismatch between perception and reality. Experimental results showed that the forces exchanged between robot and human during pHRC modulate the amplitude of PEN, representing different levels of cognitive conflict. We also found that PEN amplitude significantly decreases (p <; 0.05) when a mechanical resistance is being applied smoothly and more time in advance before an invisible obstacle, when compared to a scenario in which the resistance is applied abruptly before the obstacle. These results indicate that an earlier and smoother resistance reduces the conflict level. Consequently, this suggests that smoother changes in resistance make the interaction more intuitive.

An echinoderm can actively modulate the structural stiffness of its body wall by as much as 10 times, using the material and structural features that make up its body, including calcite ossicles, connective tissue and interossicular muscle. This capacity for variable stiffness makes it possible to adapt to the kinematics and dynamics required to perform a given task and the surrounding environment. This characteristic can improve the ability of soft material robots, which currently have limited application because of their low load-bearing capability. This paper presents a stiffness modulation method inspired by the connected ossicle structures of echinoderms. We introduce the mechanism, structure, and stiffness variation of the proposed design with respect to different ossicle shape, interval, and elastomer. Then we built a finger-shaped stiffening structure using the proposed design, measured its stiffness according to vacuum level, and showed its load-bearing capacity under control. The proposed design was then applied to a robotic gripper, a typical device that interacts with unpredictable environments and needs variable stiffening ability.

Developing robot perception systems for recognizing objects in the real world requires computer vision algorithms to be carefully scrutinized with respect to the expected operating domain. This demands large quantities of ground truth data to rigorously evaluate the performance of algorithms. This paper presents the EasyLabel tool for easily acquiring high-quality ground truth annotation of objects at pixel-level in densely cluttered scenes. In a semi-automatic process, complex scenes are incrementally built and EasyLabel exploits depth changes to extract precise object masks at each step. We use this tool to generate the Object Cluttered Indoor Dataset (OCID) that captures diverse settings of objects, background, context, sensor to scene distance, viewpoint angle and lighting conditions. OCID is used to perform a systematic comparison of existing object segmentation methods. The baseline comparison supports the need for pixel- and object-wise annotation to progress robot vision towards realistic applications. This insight reveals the usefulness of EasyLabel and OCID to better understand the challenges that robots face in the real world.

Water as a medium poses a number of challenges for robots, limiting the progress of research in underwater robotics vis-á-vis ground or aerial robotics. The primary challenges are satellite based positioning and radio communication being unusable due to high attenuation of electromagnetic waves in water. We have developed miniature, agile, easy to carry and deploy Autonomous Underwater Vehicles (AUVs) equipped with a suite of sensors for underwater environmental sensing. We previously demonstrated adaptive sampling and feature tracking, and gathered data from a lake for limnological research, with the AUV performing inertial navigation. In this paper, we demonstrate a new underwater acoustic positioning system, which allows on-board estimation of AUV position. Our system uses absolute time information from GNSS for initial clock synchronization and uses one-way-travel-time for range measurements, which makes it scalable in the number of robots. It is easily deployable and does not rely on any installed infrastructure in the environment. We describe various hardware and software components of our system, and present results from experiments in Lake Geneva.

We propose the use of Bayesian networks, which provide both a mean value and an uncertainty estimate as output, to enhance the safety of learned control policies under circumstances in which a test-time input differs significantly from the training set. Our algorithm combines reinforcement learning and end-to-end imitation learning to simultaneously learn a control policy as well as a threshold over the predictive uncertainty of the learned model, with no hand-tuning required. Corrective action, such as a return of control to the model predictive controller or human expert, is taken before the failure of tasks, when the uncertainty threshold is exceeded. We validate our method on fully-observable and vision-based partially-observable systems using cart-pole and autonomous driving simulations using deep convolutional Bayesian neural networks. We demonstrate that our method is robust to uncertainty resulting from varying system dynamics as well as from partial state observability.

This paper introduces a new designed modular tactile sensing dexterous hand for domestic service robots. This fully-actuated hand consists of 1 palm and 3 fingers, with embedded tactile sensors, motors and control boards. The palm and each finger have 2 degrees of freedom (DOFs). The modular design makes it easy to attach and detach the hand, even by inexperienced users. The tactile sensor unit with new structure can help to decrease sensor number and keep a good sensing ability. A series of experiments to test the sensor unit and evaluated the hand performance with an object set was performed in this paper. The results show that the sensor unit can provide precise sensing result and perceive continuous vibration data, and the hand has excellent grasp ability. In addition to its good performance, the hand features a cost of $500 USD with a scale of one hundred sets. This hand is affordable for researchers and for domestic service robots in the consumer market. In future research, this hand will be used to promote the robotic manipulation research based on visual and tactile data.

We present a method for EMG-driven teleoperation of non-anthropomorphic robot hands. EMG sensors are appealing as a wearable, inexpensive, and unobtrusive way to gather information about the teleoperator's hand pose. However, mapping from EMG signals to the pose space of a non-anthropomorphic hand presents multiple challenges. We present a method that first projects from forearm EMG into a subspace relevant to teleoperation. To increase robustness, we use a model which combines continuous and discrete predictors along different dimensions of this subspace. We then project from the teleoperation subspace into the pose space of the robot hand. Our method is effective and intuitive, as it enables novice users to teleoperate pick and place tasks faster and more robustly than state-of-the-art EMG teleoperation methods when applied to a non-anthropomorphic, multi-DOF robot hand.

It is nowadays well-established that locomotion can be written as a large and complex optimal control problem. Yet, current knowledge in numerical solver fails to directly solve it. A common approach is to cut the dimensionality by relying on reduced models (inverted pendulum, capture points, centroidal). However it is difficult both to account for whole-body constraints at the reduced level and also to define what is an acceptable trade-off at the whole-body level between tracking the reduced solution or searching for a new one. The main contribution of this paper is to introduce a rigorous mathematical framework based on the Alternating Direction Method of Multipliers, to enforce the consensus between the centroidal state dynamics at reduced and whole-body level. We propose an exact splitting of the whole-body optimal control problem between the centroidal dynamics (under-actuation) and the manipulator dynamics (full actuation), corresponding to a re-arrangement of the equations already stated in previous works. We then describe with details how alternating descent is a good solution to implement an effective locomotion solver. We validate this approach in simulation with walking experiments on the HRP-2 robot.

Tracking multiple prioritized tasks simultaneously with redundant robots have been investigated extensively over the last decades. Recent research focuses on combining advantages from both classical soft and strict prioritization schemes which is non-trivial. Among the proposed methods to tackle this issue, Generalized Hierarchical Control (GHC) seems to have a reasonable performance, however, it does not include a weighting matrix in the computation of the nullspace projection operator and hence cannot construct dynamically-consistent stack-of-tasks hierarchies as a special case. We extend GHC by adding dynamic-consistency to the control scheme and refer to it as DynGHC. The extension is also advantageous when choosing non-strict priorities because inertia coupling between tasks is reduced. DynGHC allows to smoothly rearrange priorities which is important for robots acting in dynamically changing contexts. Comparative simulations with a 4 DOF planar manipulator and a KUKA LWR validate our approach. Matlab and C++ source code is made available.

This article presents the design of soft material robots with tunable damping properties. This study derives from the investigation of an under-actuated dynamic approach involving multi-chamber pneumatic systems. The co-design of the mechanical parameters (stiffness and damping) of the system along with the time profile of the input allows to obtain different behaviors using a reduced number of feeding line. In this work we analyze via simulations and experiments several approaches to tune the damping of soft robots. The most effective solution employs a layer of granular material immersed in viscous oil within the chamber wall. This method has been employed to realize bending actuators with a continuous deformation pattern. Finally, we show an application involving a two-fingered gripper fed by a single pneumatic line, which is able to perform pinch and power grasp.

A physically-motivated friction model with a parametric description of the nonlinear dependency of the temperature and velocity as well as the dependency on external load is presented. The fully parametric approach extends a static friction model in the gross sliding regime. We show how it can be seamlessly integrated in standard dynamic friction models such as Lund Grenoble (LuGre) and Generalized-Maxwell-Slip (GMS). Parameters of a Harmonic Drive CSD 25 gear are experimentally identified and the final model is evaluated on a dedicated test-bed. We show the integration and effectiveness in dynamic simulation, friction compensation, and external torque estimation.