Indoor positioning systems (IPS) are essential for mobile robot navigation in environments where global positioning systems (GPS) are unavailable, such as hospitals, warehouses, and intelligent infrastructure. While current surveys may limit themselves to specific technologies or fail to provide practical application-specific details, this review summarizes IPS developments directed specifically towards mobile robotics. It examines and compares a breadth of approaches that vary across non-radio frequency, radio frequency, and hybrid sensor fusion systems, through the lens of performance metrics that include accuracy, delay, scalability, and cost. Distinctively, this work explores emerging innovations, including synthetic aperture radar (SAR), federated learning, and privacy-aware AI, which are reshaping the IPS landscape. The motivation stems from the’ increasing complexity and dynamic nature of indoor environments, where high-precision, real-time localization is essential for safety and efficiency. This literature review provides a new conceptual, cross-border pathway for research and implementation of IPS in mobile robotics, addressing both technical and application-related challenges in sectors related to healthcare, industry, and smart cities. The findings from the literature review allow early career researchers, industry knowledge workers, and stakeholders to provide secure societal, human, and economic integration of IPS with AI and IoT in safe expansions and scale-ups.

This study presents an innovative framework to improve the accessibility and usability of collaborative robot programming. Building on previous research that evaluated the feasibility of using a domain-specific language based on behaviour-driven development, this paper addresses the limitations of earlier work by integrating additional features like a drag-and-drop Blockly web interface. The system enables end users to define and execute robot actions with minimal technical knowledge, making it more adaptable and intuitive. Additionally, a gesture-recognition module facilitates multimodal interaction, allowing users to control robots through natural gestures. The system was evaluated through a user study involving participants with varying levels of professional experience and little to no programming background. Results indicate significant improvements in user satisfaction, with the system usability scale overall score increasing from 7.50 to 8.67 out of a maximum of 10 and integration ratings rising from 4.42 to 4.58 out of 5. Participants completed tasks using a manageable number of blocks (5 to 8) and reported low frustration levels (mean: 8.75 out of 100) alongside moderate mental demand (mean: 38.33 out of 100). These findings demonstrate the tool’s effectiveness in reducing cognitive load, enhancing user engagement and supporting intuitive, efficient programming of collaborative robots for industrial applications.

This research proposes an adaptive human-robot interaction (HRI) that combines voice recognition, emotional context detection, decision-making, and self-learning. The aim is to overcome challenges in dynamic and noisy environments while achieving real-time and scalable performance. The architecture is based on a three-stage HRI system: voice input acquisition, feature extraction, and adaptive decision-making. For voice recognition, modern pre-processing techniques and mel-frequency cepstral coefficients are used to robustly implement the commands. Emotional context detection is governed by neural network classification on pitch, energy, and jitter features. Decision-making uses reinforcement learning where actions are taken and then the user is prompted to provide feedback that serves as a basis for re-evaluation. Iterative self-learning mechanisms are included, thereby increasing the adaptability as stored patterns and policies are updated dynamically. The experimental results show substantial improvements in recognition accuracy along with task success rates and emotional detection. The proposed system achieved 95% accuracy and a task success rate of 96%, even against challenging noise conditions. It is apparent that emotional detection achieves a high F1-score of 92%. Real-world validation showed the system’s ability to dynamically adapt, thus mitigating 15% latency through self-learning. The proposed system has potential applications in assistive robotics, interactive learning systems, and smart environments, addressing scalability and adaptability for real-world deployment. Novel contributions to adaptive HRI arise from the integration of voice recognition, emotional context detection, and self-learning mechanisms. The findings act as a bridge between the theoretical advancements and the practical utility of further system improvements in human-robot collaboration.

This paper proposes a generalized method for designing tendon-driven serial-chained manipulators with an arbitrary number of tendon redundancy. First, a special class of tendon-driven structures is defined by introducing the controllable block triangular form (CBTF) of a null space matrix and its complementary CBTF of a structure matrix, satisfying physical constraints related to the minimal connection of tendons and to the placement of actuators. Then it is shown that any general design of tendon-driven serial manipulators can be reduced to the design of such a special class of tendon-driven structures. Two associated design problems are derived and solved. The first design problem is about finding a complementary CBTF structure matrix for a given CBTF null space matrix using algebraic relations, whereas the second one seeks the both matrices that optimize the wanted structural characteristics based on the result of the first design problem. Numerical design examples are provided to show the validity of the proposed method.

Cartographer is an algorithm that was open sourced by Google in 2016 and adapted to multiple sensors. To address issues of the original algorithm, such as the negative impact of outlier point cloud on the scan matching, and low accuracy of position fusion. This paper preprocesses the sensor data and presents HT-Carto, an improved hybrid point-cloud filtering system, and a tightly coupled LiDAR/IMU framework based on Cartographer’s front-end. The inertial measurement unit (IMU) provides initial values for the point cloud, and the IMU pre-integration combines the scan-matched pose to construct the factors, which are added as constraints to the factor graph. The result is used to update the current pose and work as odometer residuals at the back-end. The optimization of the selected strategy during point cloud preprocessing, PassThrough, and RadiusOutlierRemoval are combined to ensure quality. An actual vehicle is used in complex indoor environment to verify the stability and robustness of HT-Carto. Compared to the Cartographer, Karto, Hector, and GMapping, HT-Carto demonstrates better localization and mapping, it can obtain a more precise trajectory.

Robotic lower limb exoskeletons are wearable devices designed to augment human motor functions and enhance physical capabilities mostly adopted in healthcare and rehabilitation. The field is strongly dominated by rigid exoskeletons driven by electromagnetic actuators constituted by electrical motors, gearboxes, and cylinders. This review focuses on the design and specifications of the actuation systems of lower limb exoskeletons, with the ultimate goal of providing reporting guidelines to allow for full reproducibility. For each paper, we assessed the quality and completeness of technical characteristics with two ad hoc rating scales for motors and reducers; we extracted the main parameters of the actuation unit and a quantitative analysis of the mechanical characteristics of the individual components was carried out considering the exoskeleton application. Overall, we observed a lack of details in reporting on actuation systems equipped on exoskeletons. To overcome this limitation, herein we conclude by proposing a data form and a checklist to provide researchers with a common approach in reporting the mechanical characteristics of the actuation unit of their lower limb exoskeletons. We believe that the convergence of exoskeletons’ literature toward a clearer standardization of design and reporting will boost the development of this technology and its diffusion outside the laboratory.

Global path planning using roadmap (RM) path-planning methods including Voronoi diagram (VD), rapidly exploring random trees (RRT), and probabilistic roadmap (PRM) has gained popularity over the years in robotics. These global path-planning methods are usually combined with other path-planning techniques to achieve collision-free robot control to a specified destination. However, it is unclear which of these methods is the best choice to compute the efficient path in terms of path length, computation time, path safety, and consistency of path computation. This article reviewed and adopted a comparative research methodology to perform a comparative analysis to determine the efficiency of these methods in terms of path optimality, safety, consistency, and computation time. A hundred maps of different complexities with obstacle occupancy rates ranging from 50.95% to 78.42% were used to evaluate the performance of the RM path-planning methods. Each method demonstrated unique strengths and limitations. The study provides critical insights into their relative performance, highlighting application-specific recommendations for selecting the most suitable RM method. These findings contribute to advancing robot path-planning techniques by offering a detailed evaluation of widely adopted methods.

Walking mechanisms offer advantages over wheels or tracks for locomotion but often require complex designs. This paper presents the kinematic design and analysis of a novel overconstrained spatial a single degree-of-freedom leg mechanism for walking robots. The mechanism is generated by combining spherical four-bar linkages into two interconnecting loops, resulting in an overconstrained design with compact scalability. Kinematic analysis is applied using recurrent unit vector methods. Dimensional synthesis is performed using the Firefly optimization algorithm to achieve a near-straight trajectory during the stance phase for efficient walking. Constraints for mobility, singularity avoidance, and transmission angle are also implemented. The optimized design solution is manufactured using 3D printing and experimentally tested. Results verify the kinematic properties including near-straight-line motion during stance. The velocity profile shows low perpendicular vibrations. Advantages of the mechanism include compact scalability allowing variable stride lengths, smooth motion from overconstraint, and simplicity of a single actuator. The proposed overconstrained topology provides an effective option for the leg design of walking robots and mechanisms.

This paper introduces two enhanced control approaches to improve the performance of parallel manipulators, addressing their inherent nonlinear dynamics and complex structure. The first approach results in a hybrid control system in joint space, integrating acceleration-based control, sliding mode, and disturbance observer techniques. The control system is designed to correct tracking errors and compensate for generalized disturbances, thus improving accuracy in tracking reference positions. The second approach merges the joint-space and task-space formulations, implementing proportional-derivative controllers in task space to manage the end-effector positions while maintaining safe operational configurations. The stability of the proposed controllers is demonstrated through Lyapunov analysis, while their performance is validated through comprehensive simulations and real-time experiments.

Continuous-time batch state estimation using Gaussian processes is an efficient approach to estimate the trajectories of robots over time. In the past, relatively simple physics-motivated priors have been considered for such approaches, using assumptions such as constant velocity or acceleration. This paper presents an approach to incorporating exogenous control inputs, such as velocity or acceleration commands, into the continuous Gaussian process state estimation framework. It is shown that this approach generalizes across different domains in robotics, making it applicable to both the estimation of continuous-time trajectories for mobile robots and the estimation of quasi-static continuum-robot shapes. Results show that incorporating control inputs leads to more informed priors, potentially requiring less measurements and estimation nodes to obtain accurate estimates. This makes the approach particularly useful in situations in which limited sensing is available. For example, in a mobile robot localization experiment with sparse landmark distance measurements and frequent odometry control inputs, our approach provides accurate trajectory estimates with root-mean-square errors around 3-4 cm and 4-5 degrees, even with time intervals up to five seconds between discrete estimation nodes, which significantly reduces computation time.

In certain scenarios, the large footprint of a robot is not conducive to multi-robot cooperative operations. This paper presents a generalized single-loop parallel manipulator with remote center of motion (GSLPM-RCM), which addresses this issue by incorporating a reconfigurable base. The footprint of this RCM manipulator can be adjusted by varying the parameters of the reconfigurable base. First, utilizing configuration evolution, a reconfigurable base is constructed based on the principle of forming RCM motion. Then, according to the modular analysis method, the inverse kinematics of this parallel RCM manipulator is analyzed, and the workspace is also analyzed. Subsequently, the motion/force transmissibility of this RCM manipulator is analyzed by considering its single-loop and multi-degree of freedom characteristics. Leveraging the workspace index and transmissibility indices, dimension optimization of the manipulator is implemented. Finally, the influence of the reconfigurable base on the workspace and the transmissibility performance of the optimized manipulator is studied.

This paper proposes a mobile robot recovery mechanism for low-cost robotic systems due to vision sensor failure in vSLAM systems. The approach takes advantage of ROS architecture and adopts the Shannon Nyquist sampling theory to selectively sample path parameters that will be used for back travel in case of vision sensor failure. As opposed to point clouds normally used to store vSLAM data, this paper proposes to store and use lightweight variables namely distance between sampled points, velocity combinations, i.e., linear and angular velocity, sampled period, and yaw angle values to describe the robot path and reduce the memory space required to store these variables. In this study, low-cost robotic systems typically using cameras aided by proprioceptive sensors such as IMU during vSLAM activities are investigated. A demonstration is made on how the ROS architecture can be used in a scenario where vision sensing is adversely affected, resulting in mapping failure. Additionally, a recommendation is made for adoption of the approach for vSLAM platforms implemented on both ROS1 and ROS2. Furthermore, a proposal is made to add an additional layer to vSLAM systems that will be exclusively used for back travel in case of vision loss during vSLAM activities resulting in mapping failure.

Treating inertial measurement unit (IMU) measurements as inputs to a motion model and then preintegrating these measurements have almost become a de facto standard in many robotics applications. However, this approach has a few shortcomings. First, it conflates the IMU measurement noise with the underlying process noise. Second, it is unclear how the state will be propagated in the case of IMU measurement dropout. Third, it does not lend itself well to dealing with multiple high-rate sensors such as a lidar and an IMU or multiple asynchronous IMUs. In this paper, we compare treating an IMU as an input to a motion model against treating it as a measurement of the state in a continuous-time state estimation framework. We methodically compare the performance of these two approaches on a 1D simulation and show that they perform identically, assuming that each method’s hyperparameters have been tuned on a training set. We also provide results for our continuous-time lidar-inertial odometry in simulation and on the Newer College Dataset. In simulation, our approach exceeds the performance of an imu-as-input baseline during highly aggressive motion. On the Newer College Dataset, we demonstrate state-of-the art results. These results show that continuous-time techniques and the treatment of the IMU as a measurement of the state are promising areas of further research. Code for our lidar-inertial odometry can be found at: https://github.com/utiasASRL/steam_icp.

Automatic license plate recognition (ALPR) systems are increasingly used to solve issues related to surveillance and security. However, these systems assume constrained recognition scenarios, thereby restricting their practical use. Therefore, we address in this article the challenge of recognizing vehicle license plates (LPs) from the video feeds of a mobile security robot by proposing an efficient two-stage ALPR system. Our ALPR system combines the on-the-shelf YOLOv7x model with a novel LP recognition model, called vision transformer-based LP recognizer (ViTLPR). ViTLPR is based on the self-attention mechanism to read character sequences on LPs. To ease the deployment of our ALPR system on mobile security robots and improve its inference speed, we also propose an optimization strategy. As an additional contribution, we provide an ALPR dataset, named PGTLP-v2, collected from surveillance robots patrolling several plants. The PGTLP-v2 dataset has multiple features to cover chiefly the in-the-wild scenario. To evaluate the effectiveness of our ALPR system, experiments are carried out on the PGTLP-v2 dataset and five benchmark ALPR datasets collected from different countries. Extensive experiments demonstrate that our proposed ALPR system outperforms state-of-the-art baselines.

In numerous applications, extracting a single rotation component (termed “planar rotation”) from a 3D rotation is of significant interest. In biomechanics, for example, the analysis of joint angles within anatomical planes offers better clinical interpretability than spatial rotations. Moreover, in parallel kinematics robotic machines, unwished rotations about an axis – termed “parasitic motions” – need to be excluded. However, due to the non-Abelian nature of spatial rotations, these components cannot be extracted by simple projections as in a vector space. Despite extensive discussion in the literature about the non-uniqueness and distortion of the results due to the nonlinearity of the SO(3) group, they continue to be used due to the absence of alternatives. This paper reviews the existing methods for planar-rotation extraction from 3D rotations, showing their similarities and differences as well as inconsistencies by mathematical analysis as well as two application cases, one of them from biomechanics (flexural knee angle in the sagittal plane). Moreover, a novel, simple, and efficient method based on a pseudo-projection of the Quaternion rotation vector is introduced, which circumvents the ambiguity and distortion problems of existing approaches. In this respect, a novel method for determining the orientation of a box from camera recordings based on a two-plane projection is also proposed, which yields more precise results than the existing Perspective 3-Point Problem from the literature. This paper focuses exclusively on the case of finite rotations, as infinitesimal rotations within a single plane are non-holonomic and, through integration, produce rotation components orthogonal to the plane.

Optical microrobots are activated by a laser in a liquid medium using optical tweezers. To create visual control loops for robotic automation, this work describes a deep learning-based method for orientation estimation of optical microrobots, focusing on detecting 3-D rotational movements and localizing microrobots and trapping points (TPs). We integrated and fine-tuned You Only Look Once (YOLOv7) and Deep Simple Online Real-time Tracking (DeepSORT) algorithms, improving microrobot and TP detection accuracy by $\sim 3$% and $\sim 11$%, respectively, at the 0.95 Intersection over Union (IoU) threshold in our test set. Additionally, it increased mean average precision (mAP) by 3% at the 0.5:0.95 IoU threshold during training. Our results showed a 99% success rate in trapping events with no false-positive detection. We introduced a model that employs EfficientNet as a feature extractor combined with custom convolutional neural networks (CNNs) and feature fusion layers. To demonstrate its generalization ability, we evaluated the model on an independent in-house dataset comprising 4,757 image frames, where microrobots executed simultaneous rotations across all three axes. Our method provided mean rotation angle errors of $1.871^\circ$, $2.308^\circ$, and $2.808^\circ$ for X (yaw), Y (roll), and Z (pitch) axes, respectively. Compared to pre-trained models, our model provided the lowest error in the Y and Z axes while offering competitive results for X-axis. Finally, we demonstrated the explainability and transparency of the model’s decision-making process. Our work contributes to the field of microrobotics by providing an efficient 3-axis orientation estimation pipeline, with a clear focus on automation.

Extensible continuum robots (ECRs) offer distinct advantages over conventional continuum robots due to their ability to enhance workspace adaptability through length adjustments. This makes ECRs particularly promising for applications that require variable lengths involving the manipulation of objects in challenging environments, such as risky, cluttered, or confined. The development of ECRs necessitates careful consideration of mechanical structures, actuation methods, methods of stiffness variability, and control methods. The selection of papers is based on their relevance to ECRs within the period of 2010 to 2023 in the databases of Scopus and Web of Science. Distinguishing itself from other review papers, this paper aims to deliver a comprehensive and critical discussion about the advantages and disadvantages of ECRs concerning their mechanical structures, actuation methods, stiffness variability, and control methods. It is a beneficial resource for researchers and engineers interested in ECRs, providing essential insights to guide future developments in this field. Based on the literature, existing ECRs exhibit an inherent trade-off between flexibility and structural strength due to the absence of systematic design methods. Additionally, there is a lack of intelligent and effective controllers for achieving complex control performance and autonomous stiffness variability.

The inverse dynamics model of an industrial robot can predict and control the robot’s motion and torque output, improving its motion accuracy, efficiency, and adaptability. However, the existing inverse rigid body dynamics models still have some unmodelled residuals, and their calculation results differ significantly from the actual industrial robot conditions. The bootstrap aggregating (bagging) algorithm is combined with a long short-term memory network, the linear layer is introduced as the network optimization layer, and a compensation method of hybrid inverse dynamics model for robots based on the BLL residual prediction algorithm is proposed to meet the above needs. The BLL residual prediction algorithm framework is presented. Based on the rigid body inverse dynamics of the Newton–Euler method, the BLL residual prediction network is used to perform error compensation on the inverse dynamics model of the Franka robot. The experimental results show that the hybrid inverse dynamics model based on the BLL residual prediction algorithm can reduce the average residuals of the robot joint torque from 0.5651 N·m to 0.1096 N·m, which improves the accuracy of the inverse dynamics model compared with those of the rigid body inverse dynamics model. This study lays the foundation for performing more accurate operation tasks using industrial robots.

This paper presents the novel concept of a singularity-free tube (SFT) in the constant orientation workspace of a spatial parallel manipulator. The concept is developed and demonstrated in the context of a $6$-$6$ spatial parallel manipulator, namely, the semi-regular Stewart platform manipulator. Given two points in the said workspace, the SFT is a tubular volume which contains these points and is free of gain-type or forward-kinematic singularities. The purpose of identifying such regions in space is to allow abundant freedom to the path-planner to connect the said points by a path, which can be free of gain-type singularities simply by remaining inside the SFT at all times. To demonstrate the concept, two smooth paths obtained by formulating two different optimisation problems have been presented as examples. The SFT can be of great help in singularity-free path-planning in many similar manipulators.