IROS 2020 Papers

Autonomous valet parking is a specific application for autonomous vehicles. In this task, vehicles need to navigate in narrow, crowded and GPS-denied parking lots. Accurate localization ability is of great importance. Traditional visual-based methods suffer from tracking lost due to texture-less regions, repeated structures, and appearance changes. In this paper, we exploit robust semantic features to build the map and localize vehicles in parking lots. Semantic features contain guide signs, parking lines, speed bumps, etc, which typically appear in parking lots. Compared with traditional features, these semantic features are long-term stable and robust to the perspective and illumination change. We adopt four surround-view cameras to increase the perception range. Assisting by an IMU (Inertial Measurement Unit) and wheel encoders, the proposed system generates a global visual semantic map. This map is further used to localize vehicles at the centimeter level. We analyze the accuracy and recall of our system and compare it against other methods in real experiments. Furthermore, we demonstrate the practicability of the proposed system by the autonomous parking application.

The proliferation of mobile computing has prompted Pedestrian Dead Reckoning (PDR) to be one of the most attractive and promising indoor localization techniques for ubiquitous applications. The existing PDR approaches either suffer position drifts caused by accumulative errors or are sensitive to various users. This paper presents ARPDR, an accurate and robust PDR approach to improve the accuracy and robustness of indoor localization methods. Particularly, we propose a novel step counting algorithm based on motion models by deeply exploiting inertial sensor data. We then combine step counting with adaptive thresholding to personalize the PDR system for different users. Furthermore, we propose a novel stride-heading model with a deep neural network to predict stride lengths and walking orientations, thus the displacement errors are significantly reduced. Extensive experiments on public datasets demonstrate that ARPDR outperforms the state-of-the-art PDR methods.

In this paper, we introduce an ambiguity-aware robust active SLAM (ARAS) framework that makes use of multi-hypothesis state and map estimations to achieve better robustness. Ambiguous measurements can result in multiple probable solutions in a multi-hypothesis SLAM (MH-SLAM) system if they are temporarily unsolvable (due to insufficient information), our ARAS aims at taking all these probable estimations into account explicitly for decision making and planning, which, to the best of our knowledge, has not yet been covered by any previous active SLAM approach (which mostly consider a single hypothesis at a time). This novel ARAS framework 1) adopts local contours for efficient multi-hypothesis exploration, 2) incorporates an active loop closing module that revisits mapped areas to acquire information for hypotheses pruning to maintain the overall computational efficiency, and 3) demonstrates how to use the output target pose for path planning under the multi-hypothesis estimations. Through extensive simulations and a real-world experiment, we demonstrate that the proposed ARAS algorithm can actively map general indoor environments more robustly than a similar single-hypothesis approach in the presence of ambiguities.

We consider the problem of localization and navigation of Autonomous Underwater Vehicles (AUV) in the context of high performance subsea asset inspection missions in deep water. We propose a solution based on the recently introduced Unscented Kalman Filter on Manifolds (UKF-M) for onboard navigation to estimate the robot's location, attitude and velocity, using a precise round and rotating Earth navigation model. Our algorithm has the merit of seamlessly handling nonlinearity of attitude, and is far simpler to implement than the extended Kalman filter (EKF), which is widely used in the navigation industry. The unscented transform notably spares the user the computation of Jacobians and lends itself well to fast prototyping in the context of multi-sensor data fusion. Besides, we provide the community with feedback about implementation, and execution time is shown to be compatible with real-time. Realistic extensive Monte-Carlo simulations prove uncertainty is estimated with accuracy by the filter, and illustrate its convergence ability. Real experiments in the context of a 900m deep dive near Marseille (France) illustrate the relevance of the method.

In a collaborative scenario, robots working side by side with humans might rely on vision sensors to monitor the activity of the other agent. When occlusions of the human body occur, both the safety of the cooperation and the performance of the team can be penalized, since the robot could receive incorrect information about the ongoing cooperation. In this work, we propose a novel particle filter algorithm that, by merging the data acquired through a RGB-D camera and a MR headset, estimates online the human wrist position. This algorithm allows to significantly reduce the uncertainty of the human pose estimation, in case of both static and dynamic occlusions. To this purpose, the proposed particle filter is integrated with a detailed virtual model of the real workspace. Moreover, additional constraints describing the boundaries of the motion of the human upper body are included in a virtualized framework. The results showed that the proposed technique entails significant improvements, determining a relevant reduction of the estimation error and of the uncertainty of the estimate.

Sonoporation, which typically employs acoustic cavitation microbubbles, can enhance the permeability of the cell membrane, allowing foreign matter to enter cells across the natural barriers. However, the diameter nonuniformity and random distribution of microbubbles make it difficult to achieve controllable and high-efficiency sonoporation, while complex extern acoustic driving system also limits its applicability. Herein, we demonstrate a low-cost, expandable, and portable acoustofluidic device for cell sonoporation using acoustic streaming generated by oscillating sharp edges. The streaming-induced high shear forces can (i) quickly trap target cells at the tip of sharp edges and (ii) transiently modulate the permeability of the cell membrane, which is utilized to perform cell sonoporation events. Using our device, sonoporation is successfully achieved in a microbubble-free manner, with a sonoporation efficiency of more than 90%. Furthermore, our acoustic driving system is designed around the open-source Arduino prototyping platform due to its extendibility and portability. In addition to these benefits, our acoustofluidic device is simple to fabricate and operate, and it can work at relatively low frequency (4.6 kHz). All these advantages make our novel cell sonoporation device invaluable for many biological and biomedical applications such as drug delivery and gene transfection.

Magnetic microrobots can be propelled precisely and wirelessly in vivo using magnetic field for targeted drug delivery and early detection. They are promising for clinical trials since magnetic fields are capable of penetrating most materials with minimal interaction, and are nearly harmless to human beings. However, challenges like the biocompatibility, biodegradation and therapeutic effects of these robots must be resolved before this technique is allowed for preclinical development. In this study, we proposed a cell-robot based on macrophages for carrying drugs to kill tumors propelled by magnetic gradient-based pulling. A custom-designed system with strong gradient magnetic field system in three-dimensional (3D) space using the minimum number of coils is used for precise control of the cell-based microrobot. The cell-based microrobots were fabricated by assembling magnetic nanoparticles (Fe3O4), anti-cancer drugs (DOX) into macrophages for magnetic actuation and therapeutic effects. Vitro experiments show that cell-based microrobots can be accurately transported to the destination or approaching a targeted cancer cell. The magnetic nanoparticles have negligible effects on the cell-based microrobot and the organism, which makes the cell-based microrobot safe for in vivo experiments. The carried drugs in the cell-based microrobot can be released by the irradiation of the near-field infrared and kill the cancer cells. Further in vivo experiments prove that the cell-based microrobot can be transported to tumor area and release drugs to kill cancer effectively. The research provides biocompatible and biodegradable cell-based microrobots for early tumor prevention and targeted precision therapy.

Ensuring a safe online motion planning despite a large number of moving agents is the problem addressed in this paper. Collision avoidance is achieved without communication between the agents and without global localization system. The proposed solution is a modification of the Hybrid Reciprocal Velocity Obstacles (HRVO) combined with a tracking error estimation, in order to adapt the Velocity Obstacle paradigm to agents with kinodynamic constraints and unreliable velocity estimates. This solution, evaluated in simulation and in real test scenario with three dynamic unicycle type robots, shows an improvement over HRVO.

This paper presents an alternative impedance controller for a macro-mini robotic system in which the mini robot is unactuated. The approach is verified experimentally on a simple two-degree-of-freedom macro-mini robot. The dynamic analysis of the robot is first presented. Then, a standard impedance controller is derived and analysed. Such a controller is shown to be experimentally unstable when used with the present macro-mini mechanism. An alternative impedance controller is then proposed and analysed. While slightly more complex than the standard controller, it provides a more stable behaviour experimentally. The alternative controller also increases the effectiveness of the control by reducing the response to high-frequency motion such as hand tremor.

LiDAR is playing a more and more essential role in autonomous driving vehicles for objection detection, self localization and mapping. A single LiDAR frequently suffers from hardware failure (e.g., temporary loss of connection) due to the harsh vehicle environment (e.g., temperature, vibration, etc.), or performance degradation due to the lack of sufficient geometry features, especially for solid-state LiDARs with small field of view (FoV). To improve the system robustness and performance in self-localization and mapping, we develop a decentralized framework for simultaneous calibration, localization and mapping with multiple LiDARs. Our proposed framework is based on an extended Kalman filter (EKF), but is specially formulated for decentralized implementation. Such an implementation could potentially distribute the intensive computation among smaller computing devices or resources dedicated for each LiDAR and remove the single point of failure problem. Then this decentralized formulation is implemented on an unmanned ground vehicle (UGV) carrying 5 low-cost LiDARs and moving at 1.3m/s in urban environments. Experiment results show that the proposed method can successfully and simultaneously estimate the vehicle state (i.e., pose and velocity) and all LiDAR extrinsic parameters. The localization accuracy is up to 0.2% on the two datasets we collected. To share our findings and to make contributions to the community, meanwhile enable the readers to verify our work, we will release all our source codes1 and hardware design blueprint2 on our Github.

Over the last decades, Learning from Demonstration (LfD) has become a widely accepted solution for the problem of robot programming. According to LfD, the kinematic behavior is "taught" to the robot, based on a set of motion demonstrations performed by the human-teacher. The demonstrations can be either captured via kinesthetic teaching or external sensors, e.g., a camera. In this work, a controller for providing haptic cues of the robot's kinematic behavior to the human-teacher is proposed. Guidance is provided in procedures of kinesthetic coaching during inspection and partial modification of encoded motions. The proposed controller is based on an artificial potential field, designed to adjust the intensity of the haptic communication automatically according to the human intentions. The control scheme is proved to be passive with respect to robot's velocity and its effectiveness is experimentally evaluated in a KUKA LWR4+ robotic manipulator.

Aerial vehicles with collision resilience can operate with more confidence in environments with obstacles that are hard to detect and avoid. This paper presents the methodology used to design a collision resilient aerial vehicle with icosahedron tensegrity structure. A simplified stress analysis of the tensegrity frame under impact forces is performed to guide the selection of its components. In addition, an autonomous controller is presented to reorient the vehicle from an arbitrary orientation on the ground to help it take off. Experiments show that the vehicle can successfully reorient itself after landing upside-down and can survive collisions with speed up to 6.5m/s.

This work presents the design, fabrication, and characterization of an airflow sensor inspired by the whiskers of animals. The body of the whisker was replaced with a fin structure in order to increase the air resistance. The fin was suspended by a micro-fabricated spring system at the bottom. A permanent magnet was attached beneath the spring, and the motion of fin was captured by a readily accessible and low- cost 3D magnetic sensor located below the magnet. The sensor system was modeled in terms of the dimension parameters of fin and the spring stiffness, which were optimized to improve the performance of the sensor. The system response was then characterized using a commercial wind tunnel and the results were used for sensor calibration. The sensor was integrated into a micro aerial vehicle (MAV) and demonstrated the capability of capturing the velocity of the MAV by sensing the relative airflow during flight.

In this work, we propose a novel visuo-haptic guidance interface to enable mobile collaborative robots to follow human instructions in a way understandable by non-experts. The interface is composed of a haptic admittance module and a human visual tracking module. The haptic guidance enables an individual to guide the robot end-effector in the workspace to reach and grasp arbitrary items. The visual interface, on the other hand, uses a real-time human tracking system and enables autonomous and continuous navigation of the mobile robot towards the human, with the ability to avoid static and dynamic obstacles along its path. To ensure a safer human-robot interaction, the visual tracking goal is set outside of a certain area around the human body, entering which will switch robot behaviour to the haptic mode. The execution of the two modes is achieved by two different controllers, the mobile base admittance controller for the haptic guidance and the robot's whole-body impedance controller, that enables physically coupled and controllable locomotion and manipulation. The proposed interface is validated experimentally, where a human-guided robot performs the loading and transportation of a heavy object in a cluttered workspace, illustrating the potential of the proposed Follow-Me interface in removing the external loading from the human body in this type of repetitive industrial tasks.

This paper presents a novel control strategy for the compensation of the slippage effect during non-rigidly grasped object manipulation. A detailed dynamic model of the interconnected system composed of the robotic manipulator, the object and the internal forces and torques induced by the slippage effect is provided. Next, we design a model-based variable impedance control scheme, in order to achieve simultaneously zero convergence for the trajectory tracking error and the slippage velocity of the object. The desired damping and stiffness matrices are formulated online, by taking into account the measurement of the slippage velocity on the contact. A formal Lyapunov-based analysis guarantees the stability and convergence properties of the resulting control scheme. A set of extensive simulation studies clarifies the proposed method and verifies its efficacy.

Modular robots are autonomous systems with variable morphology, composed of independent connected computational elements, called particles or modules. Due to critical resource constraints and limited capabilities, globally unique identifier (ID) assignment to each particle is a very challenging task in modular robots. However, having a unique ID in each one remains essential for various operations and applications in this domain. For instance, it is required to establish communications between nodes and implement routing protocols. It helps in saving energy consumption and enhancing the security mechanisms. In this paper, we propose a distributed unique ID assignment method for modular robots. It is a three phases based algorithm. The first phase consists in discovering the system while building a logical tree. The second phase finds the total size of particles in the system needed for several operations in modular robots, and the third one is dedicated to the unique ID assignment. After fully optimizing the distributed algorithm, the effects of various system shapes and leader positions on the energy and time complexity are studied, while proposing fitting solutions for different requirements.

Automatic latching for charging in a disturbed environment for Unmanned Surface Vehicle (USVs) is always a challenging problem. In this paper, we propose a two-stage automatic latching system for USVs charging in berth. In Stage I, a vision-guided algorithm is developed to calculate an optimal latching position for charging. In Stage II, a novel latching mechanism is designed to compensate the movement misalignments from the water disturbance. A set of experiments have been conducted in real-world environments. The results show the latching success rate has been improved from 40% to 73.3% in the best cases with our proposed system. Furthermore, the vision-guided algorithm provides a methodology to optimize the design radius of the latching mechanism with respect to different disturbance levels accordingly. Outdoor experiments have validated the efficiency of our proposed automatic latching system. The proposed system improves the autonomy intelligence of the USVs and provides great benefits for practical applications.

This paper presents a novel approach for developing robotic grippers with variable stiffness hinges for dexterous grasps. This approach for the first time uses pneumatically actuated pouch actuators to fold and unfold morphable flaps of flexure hinges thus change stiffness of the hinge. By varying the air pressure in pouch actuators, the flexure hinge morphs into a beam with various open sections while the flaps bend, enabling stiffness variation of the flexure hinge. This design allows 3D printing of the flexure hinge using printable soft filaments. Utilizing the variable stiffness flexure hinges as the joints of robotic fingers, a light-weight and low-cost two-fingered tendon driven robotic gripper is developed. The stiffness variation caused due to the shape morphing of flexure hinges is studied by conducting static tests on fabricated hinges with different flap angles and on a flexure hinge with flaps that are bent by pouch actuators subjected to various pressures. Multiple grasp modes of the two-fingered gripper are demonstrated by grasping objects with various geometric shapes. The gripper is then integrated with a robot manipulator in a teleoperation setup for conducting a pick-and-place operation in a confined environment.

We present a path planning strategy for a magnetic millirobot where the nonlinearities in the external magnetic force field (MFF) are encoded in the graph used for planning. The strategy creates a library of candidate MFFs and characterizes their topologies by identifying the unstable manifolds in the workspace. The path planning problem is then posed as a graph search problem where the computed path consists of a sequence of unstable manifold segments and their associated MFFs. By tracking the robot's position and sequentially applying the MFFs, the robot navigates along each unstable manifold until it reaches the goal. We discuss the theoretical guarantees of the proposed strategy and experimentally validate the strategy.