Frontiers in Robotics and AI

RSS Feed for Frontiers in Robotics and AI | New and Recent Articles
Subscribe to Frontiers in Robotics and AI feed

In this paper, a distributed cooperative filtering strategy for state estimation has been developed for mobile sensor networks in a spatial–temporal varying field modeled by the advection–diffusion equation. Sensors are organized into distributed cells that resemble a mesh grid covering a spatial area, and estimation of the field value and gradient information at each cell center is obtained by running a constrained cooperative Kalman filter while incorporating the sensor measurements and information from neighboring cells. Within each cell, the finite volume method is applied to discretize and approximate the advection–diffusion equation. These approximations build the weakly coupled relationships between neighboring cells and define the constraints that the cooperative Kalman filters are subjected to. With the estimated information, a gradient-based formation control law has been developed that enables the sensor network to adjust formation size by utilizing the estimated gradient information. Convergence analysis has been conducted for both the distributed constrained cooperative Kalman filter and the formation control. Simulation results with a 9-cell 12-sensor network validate the proposed distributed filtering method and control law.

Frictionally yielding media are a particular type of non-Newtonian fluids that significantly deform under stress and do not recover their original shape. For example, mud, snow, soil, leaf litters, or sand are such substrates because they flow when stress is applied but do not bounce back when released. Some robots have been designed to move on those substrates. However, compared to moving on solid ground, significantly fewer prototypes have been developed and only a few prototypes have been demonstrated outside of the research laboratory. This paper surveys the existing biology and robotics literature to analyze principles of physics facilitating motion on yielding substrates. We categorize animal and robot locomotion based on the mechanical principles and then further on the nature of the contact: discrete contact, continuous contact above the material, or through the medium. Then, we extract different hardware solutions and motion strategies enabling different robots and animals to progress. The result reveals which design principles are more widely used and which may represent research gaps for robotics. We also discuss that higher level of abstraction helps transferring the solutions to the robotics domain also when the robot is not explicitly meant to be bio-inspired. The contribution of this paper is a review of the biology and robotics literature for identifying locomotion principles that can be applied for future robot design in yielding environments, as well as a catalog of existing solutions either in nature or man-made, to enable locomotion on yielding grounds.

In the past 2 decades, there has been increasing interest in autonomous multi-robot systems for space use. They can assemble space structures and provide services for other space assets. The utmost significance lies in the performance, stability, and robustness of these space operations. By considering system dynamics and constraints, the Model Predictive Control (MPC) framework optimizes performance. Unlike other methods, standard MPC can offer greater robustness due to its receding horizon nature. However, current literature on MPC application to space robotics primarily focuses on linear models, which is not suitable for highly non-linear multi-robot systems. Although Nonlinear MPC (NMPC) shows promise for free-floating space manipulators, current NMPC applications are limited to unconstrained non-linear systems and do not guarantee closed-loop stability. This paper introduces a novel approach to NMPC using the concept of passivity to multi-robot systems for space applications. By utilizing a passivity-based state constraint and a terminal storage function, the proposed PNMPC scheme ensures closed-loop stability and a superior performance. Therefore, this approach offers an alternative method to the control Lyapunov function for control of non-linear multi-robot space systems and applications, as stability and passivity exhibit a close relationship. Finally, this paper demonstrates that the benefits of passivity-based concepts and NMPC can be combined into a single NMPC scheme that maintains the advantages of each, including closed-loop stability through passivity and good performance through one-line optimization in NMPC.

The implementation of anthropomorphic features in regard to appearance and framing is widely supposed to increase empathy towards robots. However, recent research used mainly tasks that are rather atypical for daily human-robot interactions like sacrificing or destroying robots. The scope of the current study was to investigate the influence of anthropomorphism by design on empathy and empathic behavior in a more realistic, collaborative scenario. In this online experiment, participants collaborated either with an anthropomorphic or a technical looking robot and received either an anthropomorphic or a technical description of the respective robot. After the task completion, we investigated situational empathy by displaying a choice-scenario in which participants needed to decide whether they wanted to act empathically towards the robot (sign a petition or a guestbook for the robot) or non empathically (leave the experiment). Subsequently, the perception of and empathy towards the robot was assessed. The results revealed no significant influence of anthropomorphism on empathy and participants’ empathic behavior. However, an exploratory follow-up analysis indicates that the individual tendency to anthropomorphize might be crucial for empathy. This result strongly supports the importance to consider individual difference in human-robot interaction. Based on the exploratory analysis, we propose six items to be further investigated as empathy questionnaire in HRI.

The behaviour of shedding tears is a unique human expression of emotion. Human tears have an emotional signalling function that conveys sadness and a social signalling function that elicits support intention from others. The present study aimed to clarify whether the tears of robots have the same emotional and social signalling functions as human tears, using methods employed in previous studies conducted on human tears. Tear processing was applied to robot pictures to create pictures with and without tears, which were used as visual stimuli. In Study 1, the participants viewed pictures of robots with and without tears and rated the intensity of the emotion experienced by the robot in the picture. The results showed that adding tears to a robot’s picture significantly increased the rated intensity of sadness. Study 2 measured support intentions towards a robot by presenting a robot’s picture with a scenario. The results showed that adding tears to the robot’s picture also increased the support intentions indicating that robot tears have emotional and social signalling functions similar to those of human tears.

Image-based robot action planning is becoming an active area of research owing to recent advances in deep learning. To evaluate and execute robot actions, recently proposed approaches require the estimation of the optimal cost-minimizing path, such as the shortest distance or time, between two states. To estimate the cost, parametric models consisting of deep neural networks are widely used. However, such parametric models require large amounts of correctly labeled data to accurately estimate the cost. In real robotic tasks, collecting such data is not always feasible, and the robot itself may require collecting it. In this study, we empirically show that when a model is trained with data autonomously collected by a robot, the estimation of such parametric models could be inaccurate to perform a task. Specifically, the higher the maximum predicted distance, the more inaccurate the estimation, and the robot fails navigating in the environment. To overcome this issue, we propose an alternative metric, “task achievability” (TA), which is defined as the probability that a robot will reach a goal state within a specified number of timesteps. Compared to the training of optimal cost estimator, TA can use both optimal and non-optimal trajectories in the training dataset to train, which leads to a stable estimation. We demonstrate the effectiveness of TA through robot navigation experiments in an environment resembling a real living room. We show that TA-based navigation succeeds in navigating a robot to different target positions, even when conventional cost estimator-based navigation fails.

In this paper, the problem of attitude estimation of a quad-copter system equipped with a multi-rate camera and gyroscope sensors is addressed through extension of a sampling importance re-sampling (SIR) particle filter (PF). Attitude measurement sensors, such as cameras, usually suffer from a slow sampling rate and processing time delay compared to inertial sensors, such as gyroscopes. A discretized attitude kinematics in Euler angles is employed where the gyroscope noisy measurements are considered the model input, leading to a stochastic uncertain system model. Then, a multi-rate delayed PF is proposed so that when no camera measurement is available, the sampling part is performed only. In this case, the delayed camera measurements are used for weight computation and re-sampling. Finally, the efficiency of the proposed method is demonstrated through both numerical simulation and experimental work on the DJI Tello quad-copter system. The images captured by the camera are processed using the ORB feature extraction method and the homography method in Python-OpenCV, which is used to calculate the rotation matrix from the Tello’s image frames.

Two manipulator Jacobian matrix estimators for constrained planar snake robots are developed and tested, which enables the implementation of Jacobian-based obstacle-aided locomotion (OAL) control schemes. These schemes use obstacles in the robot’s vicinity to obtain propulsion. The devised estimators infer manipulator Jacobians for constrained planar snake robots in situations where the positions and number of surrounding obstacle constraints might change or are not precisely known. The first proposed estimator is an adaptation of contemporary research in soft robots and builds on convex optimization. The second estimator builds on the unscented Kalman filter. By simulations, we evaluate and compare the two devised algorithms in terms of their statistical performance, execution times, and robustness to measurement noise. We find that both algorithms lead to Jacobian matrix estimates that are similarly useful to predict end-effector movements. However, the unscented filter approach requires significantly lower computing resources and is not poised by convergence issues displayed by the convex optimization-based method. We foresee that the estimators may have use in other fields of research, such as soft robotics and visual servoing. The estimators may also be adapted for use in general non-planar snake robots.

Many real-world robotic applications such as search and rescue, disaster relief, and inspection operations are often set in unstructured environments with a restricted or unreliable communication infrastructure. In such environments, a multi-robot system must either be deployed to i) remain constantly connected, hence sacrificing operational efficiency or ii) allow disconnections considering when and how to regroup. In communication-restricted environments, we insist that the latter approach is desired to achieve a robust and predictable method for cooperative planning. One of the main challenges in achieving this goal is that optimal planning in partially unknown environments without communication requires an intractable sequence of possibilities. To solve this problem, we propose a novel epistemic planning approach for propagating beliefs about the system’s states during communication loss to ensure cooperative operations. Typically used for discrete multi-player games or natural language processing, epistemic planning is a powerful representation of reasoning through events, actions, and belief revisions, given new information. Most robot applications use traditional planning to interact with their immediate environment and only consider knowledge of their own state. By including an epistemic notion in planning, a robot may enact depth-of-reasoning about the system’s state, analyzing its beliefs about each robot in the system. In this method, a set of possible beliefs about other robots in the system are propagated using a Frontier-based planner to accomplish the coverage objective. As disconnections occur, each robot tracks beliefs about the system state and reasons about multiple objectives: i) coverage of the environment, ii) dissemination of new observations, and iii) possible information sharing from other robots. A task allocation optimization algorithm with gossip protocol is used in conjunction with the epistemic planning mechanism to locally optimize all three objectives, considering that in a partially unknown environment, the belief propagation may not be safe or possible to follow and that another robot may be attempting an information relay using the belief state. Results indicate that our framework performs better than the standard solution for communication restrictions and even shows similar performance to simulations with no communication limitations. Extensive experiments provide evidence of the framework’s performance in real-world scenarios.

Humans use semantic concepts such as spatial relations between objects to describe scenes and communicate tasks such as “Put the tea to the right of the cup” or “Move the plate between the fork and the spoon.” Just as children, assistive robots must be able to learn the sub-symbolic meaning of such concepts from human demonstrations and instructions. We address the problem of incrementally learning geometric models of spatial relations from few demonstrations collected online during interaction with a human. Such models enable a robot to manipulate objects in order to fulfill desired spatial relations specified by verbal instructions. At the start, we assume the robot has no geometric model of spatial relations. Given a task as above, the robot requests the user to demonstrate the task once in order to create a model from a single demonstration, leveraging cylindrical probability distribution as generative representation of spatial relations. We show how this model can be updated incrementally with each new demonstration without access to past examples in a sample-efficient way using incremental maximum likelihood estimation, and demonstrate the approach on a real humanoid robot.

The addition of geometric reconfigurability in a cable driven parallel robot (CDPR) introduces kinematic redundancies which can be exploited for manipulating structural and mechanical properties of the robot through redundancy resolution. In the event of a cable failure, a reconfigurable CDPR (rCDPR) can also realign its geometric arrangement to overcome the effects of cable failure and recover the original expected trajectory and complete the trajectory tracking task. In this paper we discuss a fault tolerant control (FTC) framework that relies on an Interactive Multiple Model (IMM) adaptive estimation filter for simultaneous fault detection and diagnosis (FDD) and task recovery. The redundancy resolution scheme for the kinematically redundant CDPR takes into account singularity avoidance, manipulability and wrench quality maximization during trajectory tracking. We further introduce a trajectory tracking methodology that enables the automatic task recovery algorithm to consistently return to the point of failure. This is particularly useful for applications where the planned trajectory is of greater importance than the goal positions, such as painting, welding or 3D printing applications. The proposed control framework is validated in simulation on a planar rCDPR with elastic cables and parameter uncertainties to introduce modeled and unmodeled dynamics in the system as it tracks a complete trajectory despite the occurrence of multiple cable failures. As cables fail one by one, the robot topology changes from an over-constrained to a fully constrained and then an under-constrained CDPR. The framework is applied with a constant-velocity kinematic feedforward controller which has the advantage of generating steady-state inputs despite dynamic oscillations during cable failures, as well as a Linear Quadratic Regulator (LQR) feedback controller to locally dampen these oscillations.

A tomographic tactile sensor based on the contact resistance of conductors is a high sensitive pressure distribution imaging method and has advantages on the flexibility and scalability of device. While the addition of internal electrodes improves the sensor’s spatial resolution, there still remain variations in resolution that depend on the contact position. In this study, we propose an optimization algorithm for electrode positions that improves entire spatial resolution by compensating for local variations in spatial resolution. Simulation results for sensors with 16 or 64 electrodes show that the proposed algorithm improves performance to 0.81 times and 0.93 times in the worst spatial resolution region of the detection area compared to equally spaced grid electrodes. The proposed methods enable tomographic tactile sensors to detect contact pressure distribution more accurately than the conventional methods, providing high-performance tactile sensing for many applications.

GNSS multipath has always been extensively researched as it is one of the hardest error sources to predict and model. External sensors are often used to remove or detect it, which transforms the process into a cumbersome data set-up. Thus, we decided to only use GNSS correlator outputs to detect large amplitude multipath, on Galileo E1-B and GPS L1 C/A, using a convolutional neural network. This network was trained using 101 correlator outputs being used as a theoretical classifier. To take advantage of the strengths of convolutional neural networks for image detection, images representing the correlator output values as a function of delay and time were generated. The presented model has a F score of 94.7% on Galileo E1-B and 91.6% on GPS L1 C/A. To reduce the computational load, the number of correlator outputs and cor- relator sampling frequency was then decreased by a factor of 4, and the convolutional neural network still has a F score of 91.8% on Galileo E1-B and 90.5% on GPS L1 C/A in that case.

HIRO is a health-assistive robot deployed in an outpatient primary care clinic to sanitize its premises, monitor people in its proximity for their temperature and donning of masks, and to usher them to service points. This study aimed to determine the acceptability, perceptions of safety and concerns among the patients, visitors and the polyclinic healthcare workers (HCW) on HIRO. A cross sectional questionnaire survey was conducted from March to April 2022 when HIRO was at Tampines Polyclinic in eastern Singapore. 170 multidisciplinary HCW serve approximately 1000 patients and visitors daily at this polyclinic. The sample size of 385 was computed using proportion of 0.5, 5% precision and 95% confidence interval. Research assistants administered an e-survey to gather demographic data and feedback from 300 patients/visitors and 85 HCW on their perceptions of HIRO using Likert scales. The participants watched a video on HIRO’s functionalities and were given the opportunity to directly interact with it. Descriptive statistics were performed and figures were presented in frequencies and percentages. The majority of participants viewed HIRO’s functionalities favourably: sanitizing (96.7%/91.2%); checking proper mask-donning (97%/89.4%); temperature-monitoring (97%/91.7%); ushering (91.7%/81.1%); perceived user-friendliness (93%/88.3%) and improvement in clinic experience (96%/94.2%). Minority of participants perceived harm from liquid disinfectant (29.6%/31.5%) and that its voice-annotated instructions may be upsetting (14%/24.8%). Most participants accepted HIRO’s deployment at the polyclinic and perceived it to be safe. HIRO used ultraviolet irradiation for sanitization during after-clinic hours instead of disinfectant due to its perceived harm.

Combining and completing point cloud data from two or more sensors with arbitrarily relative perspectives in a dynamic, cluttered, and complex environment is challenging, especially when the two sensors have significant perspective differences while the large overlap ratio and feature-rich scene cannot be guaranteed. We create a novel approach targeting this challenging scenario by registering two camera captures in a time series with unknown perspectives and human movements to easily use our system in a real-life scene. In our approach, we first reduce the six unknowns of 3D point cloud completion to three by aligning the ground planes found by our previous perspective-independent 3D ground plane estimation algorithm. Subsequently, we use a histogram-based approach to identify and extract all the humans from each frame generating a three-dimensional (3D) human walking sequence in a time series. To enhance accuracy and performance, we convert 3D human walking sequences to lines by calculating the center of mass (CoM) point of each human body and connecting them. Finally, we match the walking paths in different data trials by minimizing the Fréchet distance between two walking paths and using 2D iterative closest point (ICP) to find the remaining three unknowns in the overall transformation matrix for the final alignment. Using this approach, we can successfully register the corresponding walking path of the human between the two cameras’ captures and estimate the transformation matrix between the two sensors.

The field of multi-robot systems (MRS) has recently been gaining increasing popularity among various research groups, practitioners, and a wide range of industries. Compared to single-robot systems, multi-robot systems are able to perform tasks more efficiently or accomplish objectives that are simply not feasible with a single unit. This makes such multi-robot systems ideal candidates for carrying out distributed tasks in large environments—e.g., performing object retrieval, mapping, or surveillance. However, the traditional approach to multi-robot systems using global planning and centralized operation is, in general, ill-suited for fulfilling tasks in unstructured and dynamic environments. Swarming multi-robot systems have been proposed to deal with such steep challenges, primarily owing to its adaptivity. These qualities are expressed by the system’s ability to learn or change its behavior in response to new and/or evolving operating conditions. Given its importance, in this perspective, we focus on the critical importance of adaptivity for effective multi-robot system swarming and use it as the basis for defining, and potentially quantifying, swarm intelligence. In addition, we highlight the importance of establishing a suite of benchmark tests to measure a swarm’s level of adaptivity. We believe that a focus on achieving increased levels of swarm intelligence through the focus on adaptivity will further be able to elevate the field of swarm robotics.

Soft pneumatic artificial muscles are increasingly popular in the field of soft robotics due to their light-weight, complex motions, and safe interfacing with humans. In this paper, we present a Vacuum-Powered Artificial Muscle (VPAM) with an adjustable operating length that offers adaptability throughout its use, particularly in settings with variable workspaces. To achieve the adjustable operating length, we designed the VPAM with a modular structure consisting of cells that can be clipped in a collapsed state and unclipped as desired. We then conducted a case study in infant physical therapy to demonstrate the capabilities of our actuator. We developed a dynamic model of the device and a model-informed open-loop control system, and validated their accuracy in a simulated patient setup. Our results showed that the VPAM maintains its performance as it grows. This is crucial in applications such as infant physical therapy where the device must adapt to the growth of the patient during a 6-month treatment regime without actuator replacement. The ability to adjust the length of the VPAM on demand offers a significant advantage over traditional fixed-length actuators, making it a promising solution for soft robotics. This actuator has potential for various applications that can leverage on demand expansion and shrinking, including exoskeletons, wearable devices, medical robots, and exploration robots.