Most research on UAV swarm architectures remains confined to simulation-based studies, with limited real-world implementation and validation. In order to mitigate this issue, this research presents an improved task allocation and formation control system within ARCog-NET (Aerial Robot Cognitive Architecture), aimed at deploying autonomous UAV swarms as a unified and scalable solution. The proposed architecture integrates perception, planning, decision-making, and adaptive learning, enabling UAV swarms to dynamically adjust path planning, task allocation, and formation control in response to evolving mission demands. Inspired by artificial intelligence and cognitive science, ARCog-NET employs an Edge-Fog-Cloud (EFC) computing model, where edge UAVs handle real-time data acquisition and local processing, fog nodes coordinate intermediate control, and cloud servers manage complex computations, storage, and human supervision. This hierarchical structure balances real-time autonomy at the UAV level with high-level optimization and decision-making, creating a collective intelligence system that automatically fine-tunes decision parameters based on configurable triggers. To validate ARCog-NET, a realistic simulation framework was developed using SITL (Software-In-The-Loop) with actual flight controller firmware and ROS-based middleware, enabling high-fidelity emulation. This framework bridges the gap between virtual simulations and real-world deployments, allowing evaluation of performance in environmental monitoring, search and rescue, and emergency communication network deployment. Results demonstrate superior energy efficiency, adaptability, and operational effectiveness compared to conventional robotic swarm methodologies. By dynamically optimizing data processing based on task urgency, resource availability, and network conditions, ARCog-NET bridges the gap between theoretical swarm intelligence models and real-world UAV applications, paving the way for future large-scale deployments.

The robot manipulator is commonly employed in the space station experiment cabinet for the disinfection task. The challenge lies in devising a motion trajectory for the robot manipulator that satisfies both performance criteria and constraints within the confined space of an experimental cabinet. To address this issue, this paper proposes a trajectory planning method in joint space. This method constructs the optimal trajectory by transforming the original problem into a constrained multi-objective optimization problem. This is then solved and integrated with the seventh-degree B-spline curve. The optimization algorithm utilizes an indicator-based adaptive differential evolution algorithm, enhanced with improved Tent chaotic mapping and opposition-based learning for population initialization. The method employed the Fréchet distance to design a trajectory selection strategy based on the Pareto solutions to ensure that the planned trajectory complies with Cartesian space requirements. This allows the robot manipulator end-effector to approximate the desired path in Cartesian space closely. The findings indicate that the proposed method can effectively design the robot manipulator trajectory, considering both joint motion performance and end-effector motion constraints. This ensures that the robot manipulator operates efficiently and safely within the experimental cabinet.

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.

In this work, the problem of reliably checking collisions between robot manipulators and the surrounding environment in short time for tasks, such as replanning and object grasping in clutter, is addressed. Geometric approaches are usually applied in this context; however, they can result not suitable in highly time-constrained applications. The purpose of this paper is to present a learning-based method able to outperform geometric approaches in clutter. The proposed approach uses a neural network (NN) to detect collisions online by performing a classification task on the input represented by the depth image or point cloud containing the robot gripper projected into the application scene. Specifically, several state-of-the-art NN architectures are considered, along with some customization to tackle the problem at hand. These approaches are compared to identify the model that achieves the highest accuracy while containing the computational burden. The analysis shows the feasibility of the robot collision checker based on a deep learning approach. In fact, such approach presents a low collision detection time, of the order of milliseconds on the selected hardware, with acceptable accuracy. Furthermore, the computational burden is compared with state-of-the-art geometric techniques. The entire work is based on an industrial case study involving a KUKA Agilus industrial robot manipulator at the Technology $\&$ Innovation Center of KUKA Deutschland GmbH, Germany. Further validation is performed with the Amazon Robotic Manipulation Benchmark (ARMBench) dataset as well, in order to corroborate the reported findings.

This paper proposes a LiDAR-inertial odometry (LIO) based on the dynamic voxel merging and smoothing method, DV-LIO. In this approach, a local map management mechanism based on feature distribution is introduced to unify the features of similar adjacent voxels through dynamic merging and segmentation, thereby improving the perceptual consistency of environmental features. Moreover, a novel noise detector that performs noise detection and incremental filtering by evaluating the consistency of voxel features is designed to further reduce local map noise and improve mapping accuracy while ensuring real-time algorithm performance. Meanwhile, to ensure the computational efficiency of the LIO system, a point cache is set for each voxel, which allows the voxel to be updated incrementally and intermittently. The proposed method is extensively evaluated on datasets gathered over various environments, including campus, park, and unstructured gardens.

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.

This paper presents an efficient trajectory planning method for a 4-DOF robotic arm designed for pick-and-place manipulation tasks. The method addresses several challenges, where traditional optimization approaches struggle with high dimensionality, and data-driven methods are costly to collect enough data. The proposed approach leverages Bézier curves for computationally efficient, smooth trajectory generation, minimizing abrupt changes in motion. When continuous solutions for the end-effector angle are unavailable, joint angles are interpolated using Bézier or Hermite interpolation. Additionally, we use custom metrics to evaluate deviation between the interpolated trajectory and the original trajectory, as well as the overall smoothness of the path. When a continuous solution exists, the trajectory is treated as a Gaussian process, where a prior factor is generated using the centerline. This prior is then combined with a smoothness factor to optimize the trajectory, ensuring it remains as smooth as possible within the feasible solution space through stochastic gradient descent. The method is evaluated through simulations in Nvidia Isaac Sim; results highlight the method’s suitability, and future work will explore enhancements in prior trajectory integration and smoothing techniques.

Continuum robot-based surgical systems are becoming an effective tool for minimally invasive surgery. A flexible, dexterous, and compact robot structure is suitable for carrying out complex surgical operations. In this paper, we propose performance metrics for dexterity based on data density. Data density at a point in the workspace is higher if the number of reachable points is higher, with a unique configuration lying in a small square box around a point. The computation of these metrics is performed with forward kinematics using the Monte Carlo method and, hence, is computationally efficient. The data density at a particular point is a measure of dexterity at that point. In contrast, the dexterity distribution property index is a measure of how well dexterity is distributed across the workspace according to desired criteria. We compare the dexterity distribution property index across the workspace with the dexterity index based on the dexterous solid angle and manipulability-based approach. A comparative study reveals that the proposed method is simple and straightforward because it uses only the position of the reachable point as the input parameter. The method can quantify and compare the performance of different geometric designs of hyper-redundant and multisegment continuum robots based on dexterity.

In response to the auxiliary requirements for the treatment and prevention of lumbar diseases, based on the biomechanical characteristics of the human waist, a novel unpowered rigid-flexible coupling waist exoskeleton with multiple degrees of freedom and its human-exoskeleton parallel wearable equivalent research prototype are proposed, further focusing on the encompassing kinematic compatibility and dynamic load-bearing effectiveness of the biomimetic coordination, an in-depth analysis is performed on the multi-body dynamic dimensional synthesis and its methodological research. Initially, based on the rigid-flexible coupling characteristics and experimental biomechanical data of the lumbar region in the sagittal plane, an accurate multi-body system dynamics model of the research prototype, which incorporates the rigid-flexible coupling characteristics, is systematically constructed. Subsequently, to effectively quantify the biomimetic coordination of the exoskeleton, a novel comprehensive optimization index, termed biomimetic load-bearing comfort, is proposed. Finally, by utilizing this index, the exoskeleton is optimized in dimension by employing a thorough combination of multi-dimensional spatial search algorithm and compression factor particle swarm algorithm. The simulation results validate the correctness and effectiveness of both the dynamic dimensional synthesis and its methodology. Furthermore, the study also reveals that the optimized exoskeleton’s passive working mode showcases favorable biomimetic coordination. These results are crucial for progressing the research on the biomimetic load-bearing capacities of other exoskeletons.

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.

This paper introduces the Arachne System, a scalable, cost-effective mobile microrobot swarm platform designed for educational and research applications. It details the design, functionalities, and potential of the Arachne Bots, emphasizing their accessibility to users with minimal robotics expertise. By providing a comprehensive overview of the system’s hardware, sensory capabilities, and control algorithms, the paper demonstrates the platform’s capacity to democratize and reduce entry barriers in mobile robotic swarms research, fostering innovation and educational opportunities in the field. Extensive experimental validation of the system showcases its broad range of capabilities and effectiveness in real-world implementation.

This paper presents a climbing robot (CR) designed for the purpose of pipeline maintenance, with capability to avoid the risks inherent in manual operations. In the design process, a three degree of freedom (DOF) parallel mechanism coupled with a remote center of motion (RCM) mechanism linkage mechanism were designed to serve as the CR’s climbing mechanism, which met the specific demands for climbing movements. The modified Kutzbach–Grübler formula and the screw theory were applied to calculate the DOFs of the CR. Then, the inverse and forward position analysis for the CR was derived. Furthermore, velocity and acceleration analysis of parallel mechanism were conducted and derived the Jacobian matrix, through which the singularity of parallel mechanism was analyzed. In order to evaluate kinematic performance of parallel mechanism, the motion/force transmission index (LTI) of workspace was calculated, which directed the followed dimensional optimization process. According to the optimization result, a prototype was constructed and a series of motion experiments were carried out to validate its climbing capability.

A modeling method of multi-objective optimization design for parallel mechanisms (PMs) is proposed, whose implementation is illustrated with 2RPU-RPS mechanism as an example. The orientation of biased output axis on moving platform is depicted by spherical attitude angles, and its kinematic model is deduced through vector method. With screw theory as mathematic tool, a comprehensive evaluation method of kinematic performance for PM is established. On this basis, the expensive constrained multi-objective optimization model of dimensional parameters for the discussed mechanism is constructed. The NSDE-II algorithm, formed by replacing the genetic algorithm operators in non-dominated sorting genetic algorithm II (NSGA-II) with DE operators, is utilized to solve this multi-objective optimization problem, thus obtaining multiple Pareto optimal solutions with engineering application significance, which proves the feasibility and effectiveness of the proposed modeling method and algorithm. Moreover, the normalization coverage space and the minimum adjacent vector angle are proposed to evaluate the computational performance of NSDE-II. Finally, the potential engineering application value for the optimized 2RPU-RPS PM is presented.

Screw theory serves as an influential mathematical tool, significantly contributing to mechanical engineering, with particular relevance to mechanism science and robotics. The instantaneous screw and the finite displacement screw have been used to analyse the degree of freedom and perform kinematic analysis of linkage mechanisms with only lower pairs. However, they are not suitable for higher pair mechanisms, which can achieve complex motions with a more concise structure by reasonably designing contact contours, and they possess advantages in some particular areas. Therefore, to improve the adaptability of screw theory, this paper aims to analyse higher kinematic pair (HKP) mechanisms and proposes a method to extend instantaneous screw and finite displacement screw theory. This method can not only analyse the instantaneous degree of freedom of HKP mechanisms but also determine the relationships between the motion variables of HKP mechanisms. Furthermore, this method is applied to calculate the degree of freedom and the relationships between the motion angles in both planar and spatial cam mechanisms, thereby demonstrating its efficiency and advantages.

Upper extremity rehabilitation robots have become crucial in stroke rehabilitation due to their high durability, repeatability, and task-specific capabilities. A significant challenge in assessing the comfort performance of these robots is accurately calculating the human-robot interaction forces. In this study, a four-degree-of-freedom (4-DOF) upper extremity rehabilitation robot mechanism, kinematically compatible with the human upper limb, is proposed. Based on this mechanism, an algorithm for estimating human-robot interaction forces is developed using Newton-Euler dynamics. A prototype of the proposed robot is constructed, and a series of comparative experiments are carried out to validate the feasibility of the proposed force estimation approach. The results indicate that the proposed method reliably predicts interaction forces with minimal deviation from experimental data, demonstrating its potential for application in upper limb rehabilitation robots. This work provides a foundation for future studies focused on comfort evaluation and optimization of rehabilitation robots, with significant practical implications for improving patient rehabilitation outcomes.

This paper deals with the optimization of a new redundant spherical parallel manipulator (New SPM). This manipulator consists of two spherical five-bar mechanisms connected by the end-effector, providing three degrees of freedom, and has an unlimited self-rotation capability. Three optimization procedures based on the genetic algorithm method were carried out to improve the dexterity of the New SPM. The first and the second optimizations were applied to a symmetric New SPM structure, while the third was applied to an asymmetric New SPM structure. In both cases, the optimization was performed using an objective function defined by the quadratic sum of link angles. In addition, certain criteria and constraints were implemented. The obtained results demonstrate significant improvements in the dexterity of the New SPM and its capability of an unlimited self-rotate in an extended workspace. A comparison of the self-rotation performances between the classical 3-RRR SPM (R for revolute joint) and the New SPM is also presented.

In response to the prevailing trend of an aging society and the increasing requirements of rehabilitation, this paper presents an approach involving brain-machine interaction (BMI) for a single-degree-of-freedom (1-DOF) sit-to-stand transfer robot. Based on a 1-DOF rehabilitation robot, three experiment paradigms involving motor imagery (MI), action observation of motor imagery (AO-MI) and motor execution are designed using both electroencephalography (EEG) and electromyography (EMG). To enhance motion intention recognition accuracy, a Gumbel-ResNet-KANs decoding model is established. The Gumbel-ResNet-KANs model integrates the Gumbel-Softmax method with the ResNet-KANs network module and demonstrates strong decoding capability, as demonstrated by comparative tests in this paper. To validate the effect of robotic assistance, EEG and EMG coherence are analyzed to assess the impact of robotic assistance on rehabilitation from a neuromuscular perspective in both assisted and unassisted conditions. We assessed the effect of robotics on rehabilitation from an emotional perspective by analyzing the difference between the differential entropy of the right and left brain. The proposed study also reveals that the movement-related cortical potentials in AO-MI are beneficial for promoting the performance of BMI in sit-to-stand training, which provides a possible approach for the development of new types of robots for lower limb rehabilitation.

The path navigation of robot in an entirely known space is presented by various researchers in the recent times. The navigational complexity arises when a robot moves in a completely unknown and complex environment from one defined start to a designated desired location. As the success of the nature-inspired algorithms in the unclear navigational problem is better, therefore, an improved butterfly optimization algorithm (IBOA) to determine the optimal feasible path for a humanoid robot navigating through a platform cluttered with both known and unfamiliar barriers is presented in this study. The BOA is inspired by the food-gathering habits of butterflies, where the sense of smell is the vital parameter in the global optimal search. However, the performance of this technique in the complex environment is poor, as a result, the chances of being trapped in local minima are more. Hence, the BOA is improved by using a nonlinear weight reduction strategy in updating the position of the butterflies in every iteration. The simulation is carried out in the Webots platform by considering variable-legged robot, NAO, in an unfamiliar environment. The outcomes derived from the simulation and real assessments demonstrate the potential of the proposed technique and compare with other existing algorithms, which highlights the potential and efficacy of the proposed IBOA algorithm.

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.

This paper focuses on the feature-based visual-inertial odometry (VIO) in dynamic illumination environments. While the performance of most existing feature-based VIO methods is degraded by the dynamic illumination, which leads to unstable feature association, we propose a tightly-coupled VIO algorithm termed RAFT-VINS, integrating a Lite-RAFT tracker into the visual inertial navigation system (VINS). The key module of this odometry algorithm is a lightweight optical flow network designed for accurate feature tracking with real-time operation. It guarantees robust feature association in dynamic illumination environments and thereby ensures the performance of the odometry. Besides, to further improve the accuracy of the pose estimation, a moving consistency check strategy is developed in RAFT-VINS to identify and remove the outlier feature points. Meanwhile, a tightly-coupled optimization-based framework is employed to fuse IMU and visual measurements in the sliding window for efficient and accurate pose estimation. Through comprehensive experiments in the public datasets and real-world scenarios, the proposed RAFT-VINS is validated for its capacity to provide trustable pose estimates in challenging dynamic illumination environments. Our codes are open-sourced on https://github.com/USTC-AIS-Lab/RAFT-VINS.