This paper introduces a novel Gramian-based quantitative metric to evaluate the disturbance rejection capabilities of linear unstable systems. The proposed metric addresses key limitations of the previously introduced degree of disturbance rejection (DoDR) metrics, including their dependency on the final time and numerical problems arising from differential equation computations. Specifically, this study defines the steady-state solution of the DoDR metric, which avoids numerical issues by relying only on solving four algebraic equations, even when the Gramian matrices diverge. This study further strengthens its contributions by providing rigorous mathematical proofs supporting the proposed method, ensuring a strong theoretical foundation. The derived results demonstrate that the proposed metric represents the sum of the steady-state input energies required to reject the disturbances in the asymptotically stable and anti-stable subsystems. Numerical examples demonstrated that the proposed metric maintained the physical meaning of the original DoDR while offering practical computational advantages. This study represents a significant step toward the efficient and reliable assessment of disturbance rejection capabilities in unstable systems.

In-hand manipulation with Cartesian-force-control-based pushing primitives is introduced to achieve the precise placement of an object in a desired position at a manufacturing site. In the bin picking process, achieving the desired grasping posture is challenging due to limitations in the sensing and control of the robotic arm, interference from clustered objects, and unintended collisions, which hinder achieving the planned pose. Even under such conditions, in cases that require precise operations, such as manufacturing processes, maintaining a desired placement posture is crucial for the precise placement of objects into the machine slot. In this paper, a pushing primitive incorporating force feedback control is applied to ensure that the gripper is consistently positioned at the edge of the grasped object regardless of the initial grasping position by utilizing the surrounding environment of the processing machine. Modeling the exact contact friction between the gripper and the grasped object is challenging; therefore, instead of relying on a motion planning approach, we addressed the problem using a control method that leverages feedback from the external force information of the robot manipulator. Additional sensors such as external cameras or tactile sensors in the gripper are not required. The pushing primitive is executed by applying a force greater than the frictional force between the gripper and the grasped object, leveraging the surrounding environment. Experimental verification confirmed that the proposed method achieves precise placement into the machine slot, regardless of initial grasping positions. It also proved to be effective on an actual testbed.

Accurate parameter estimation of unknown objects is crucial for the precise and safe manipulation of robotic systems in applications such as object positioning, assembly, and collaborative manipulation tasks involving humans or multiple robot agents. However, measurement data obtained from sensors often contain uncertainties, making accurate parameter estimation challenging. In this paper, a systematic methodology for unmodeled dynamics identification that represents uncertainties in measured sensor data is proposed for accurate online parameter estimation of task objects. The sparse identification of nonlinear dynamics (SINDy) technique, a recent machine learning approach, is employed to identify unmodeled dynamics. First, in the learning process, an unmodeled dynamic equation can be obtained by establishing residual data, which are obtained by subtracting the dynamics of prior known objects from measured sensor data and by designing proper candidates that successfully capture uncertain behavior. Second, in the online parameter estimation process for an unknown object, estimation results that are not contaminated by uncertainties can be obtained by incorporating the identified unmodeled dynamic equation into the nominal object equation. To verify the robustness and estimation accuracy of the proposed methodology, experiments were conducted using various objects. The experiments demonstrate that the proposed method improves the estimation accuracy by reducing errors by 15.71

A novel design of a magnetic levitation (maglev) ropeless elevator for semiconductor wafer vertical transport is newly presented. To satisfy high cleanliness during vertical transport of the wafer, linear motor lifting and maglev guiding are desirable instead of the conventional rope lifting and wheel-based guiding method. Owing to the physically noncontact maglev guiding, particle-free and high-speed operation can be achieved with remarkable ride quality. In this paper, practical challenges for wafer transport, such as the eccentricity of the mass center from the actuating axis, severe acceleration/deceleration conditions as well as periodic large disturbances from normal forces of the linear motor, resonance of the elevator, and manufacturing tolerance of the guide rail, are further considered to satisfy the ride performance in wafer transport by designing a robust feedback controller with a loop shaping technique. A full-scale maglev ropeless elevator was constructed to experimentally validate the effectiveness of the proposed method. The experimental results demonstrated excellent magnetic levitation guiding performance in 5 degrees of freedom, with a maximum airgap fluctuation of 227 mu m under harsh lifting conditions V-max =2600mm/s).

Mobile robots that are deployed both indoors and outdoors require the capability of recognizing their environment in real-time to improve their autonomous navigation. Many researchers study the leverage of cameras and light detection and ranging (LiDAR) sensors in combination to generate a representation of the environment. LiDAR is widely adopted for its ability to create high-precision maps in mobile robots and autonomous vehicles. With advanced deep learning techniques, various camera perception methods such as semantic segmentation, object detection, and classification represented remarkable performance. In this paper, we propose an efficient traversability map that fuses the recognition capability of cameras with the accurate mapping of 3D (LiDAR) sensors. Finally, we show an approximate F1 score performance of 0.83 compared to the LiDAR-based map generation method.

Magnetic levitation can reduce particulate contamination that occurs during wafer transportation in the semiconductor manufacturing process. This technology radically eliminates contact between the wafer and the transport system, reducing friction, wear, and particle generation. Therefore, it is suitable for achieving high cleanliness in the ultra-fine line-width semiconductor production process and solving the need for particle removal in a vacuum environment. In this study, the roller and linear motion guide components of the wafer transfer system were replaced with a magnetic levitation module, and a robot arm was installed on top to transport a single wafer. A posture controller and a current controller were designed, and test equipment simulating the wafer transfer system was also manufactured and tested. Regarding mover and system identification, a sine sweep test was performed on the motion axis of the five degrees of freedom. Through the obtained system identification, it was possible to design the posture controller more precisely. Moreover, through levitation in standstill experiments and high-speed operation experiments, the wafer transport system can be used to verify dust-free high-speed transport and accurate positioning performance.

This paper and the accompanying demonstration video show our use case of WALL-ET, a social, cognitive, mobile robot platform with a height-adjustable table unit, which assists workers in warehouses or supermarkets when performing unergonomic tasks. With the help of Augmented Reality-glasses and a camera-based activity detection, the system can infer a worker’s intention of lifting boxes from a specified shelf region.

A multiple-actuator fault isolation approach for overactuated electric vehicles (EVs) is designed with a minimal ℓ1-norm solution. As the numbers of driving motors and steering actuators increase beyond the number of controlled variables, an EV becomes an overactuated system, which exhibits actuator redundancy and enables the possibility of fault-tolerant control (FTC). On the other hand, an increase in the number of actuators also increases the possibility of simultaneously occurring multiple faults. To ensure EV reliability while driving, exact and fast fault isolation is required; however, the existing fault isolation methods demand high computational power or complicated procedures because the overactuated systems have many actuators, and the number of simultaneous fault occurrences is increased. The method proposed in this paper exploits the concept of sparsity. The underdetermined linear system is defined from the parity equation, and fault isolation is achieved by obtaining the sparsest nonzero component of the residuals from the minimal ℓ1-norm solution. Therefore, the locations of the faults can be obtained in a sequence, and only a consistently low computational load is required regardless of the isolated number of faults. The experimental results obtained with a scaled-down overactuated EV support the effectiveness of the proposed method, and a quantitative index of the sparsity condition for the target EV is discussed with a CarSim-connected MATLAB/Simulink simulation.

An omnidirectional mobile robot with lift mechanism has been developed to assist worker to transport heavy goods placed on high position. However, as the center of gravity heightens and the stiffness decreases with introducing the lift mechanism, and the suspension is adopted for consistent contact between mecanum wheel and the ground, a large amount of vertical vibration is inevitable. In this paper, the simultaneous controller for the driving velocity tracking and vibration reduction without additional actuators is developed based on a combined model with longitudinal and vertical motion, i.e. a suspended cart-pole inverted pendulum model. Proper Filters were designed to effectively remove uninterested characteristics such as mecanum wheel roller contact vibration and static inclination so that the performance of the vibration reduction controller can be optimized. From the experimental results, the performance of the proposed method is verified that the magnitude and time of residual vibration greatly reduced.

Autonomous mobile robots equipped with long-stroke lift modules have been developed to assist human workers to transport high positioned heavy objects in the field of logistics. A large amount of rolling vibration during acceleration or deceleration when driving is inevitable for a robot with a high center of mass and low stiffness mechanism. In this study, the simultaneous control of reference velocity tracking and vibration reduction, only using driving motors, is proposed based on a suspended cart-pole inverted pendulum model that combines velocity and vertical vibration motion. A model predictive controller was adopted to account for constraints that prevented the object from slipping when placed atop the lift by saturating the magnitude of the rolling angular acceleration. The simulation results verified that the residual vibration time and magnitude significantly decreased with the proposed controller compared to the controller accounting for driving velocity alone. Additionally, object slippage prevention was ensured with taking minimal loss of the reference velocity tracking performance.

The experimental verification of the optimal input design method for fault identification is performed using a scaled-down overactuated electric vehicle. In the previous study, online fault identification was achieved by utilizing all the characteristics of an overactuated system (Park and Park, 2016). The perturbation input signal for the actuator fault identification can be applied to the faulty actuators to suppress most of the control performance loss. The scaled-down vehicle contains four independent driving motors and four independent wheel steering motors to model an extremely overactuated system. The lateral velocity and yaw rate are estimated using the state observer to realize feedback control, and the cornering stiffness is determined based on the estimated values. Experimental verification is performed using steady state cornering maneuvers with sudden actuator faults. The experiments with the scaled-down vehicle support the performance of the optimal input design method. When sensor noise and modeling uncertainties exist, the results from our method were much more precise than the results obtained using the conventional white noise perturbation input signal.

In recent years, with increasing levels of e-commerce, the use of mobile robots in logistics has increased significantly. The most common driving mechanism for mobile robots is the differential type, but the omnidirectional type mechanism, too, is widely used due to its high maneuverability in crowded environments. Meanwhile, studies on a variable footprint mechanism (VFM) have been conducted to improve the maneuverability and driving stability of mobile robots. The extant VFMs offer some advantages, but they also have some limitations such as limited motion range due to structural singularity and complex controller design due to highly coupled parameters in kinematics. In this paper, an omnidirectional variable footprint mechanism (OVFM) is proposed to overcome these limitations. The proposed OVFM is designed with a diamond-shaped link structure that maximizes the motion range without singularity and a synchronized drive chain to achieve variable motion. The kinematic model of the proposed OVFM is mathematically derived, and it is demonstrated that the parameters related to VFM are only coupled to the rotation of the mobile robot, which could simplify controller design. The kinematic model is verified using the dynamics simulation tool RecurDyn in two different scenarios (one for variable motion under the static condition and the other for variable motion under the dynamic condition). The results indicate that the proposed kinematic model can be used to predict the motion of OVFM with acceptable errors that originate from the inherent nature of slip and vibration of Mecanum wheels.

To cope with a decrease in the labor force, a change in the production environment, and demand for productivity improvement in the industrial field, the operating system needs to be upgraded by combining artificial intelligence/big data/IoT technology and production automation using industrial robots. There is a growing demand for a system capable of working with people in non-structured environments. In this paper, we introduce a multifunctional autonomous mobile robot that can handle various tasks in non-structured environments. It includes the design of a modular mobile manipulator that can be reconfigured according to the payload or the type of movement, AI-based pick-and-place, obstacle avoidance, force control-based manipulation, and driving stability check method.

With the increasing worldwide demand to automate product distribution, technology firms are introducing various mobile robots to the market. Generally, the maneuverability of the mobile robot is determined by the type and location of its wheels. Omnidirectional mobile platforms suitable for the agile motion required during instantaneous obstacle avoidance are widely used in congested environments like indoor factories. Previous omnidirectional mobile robots have been developed for specific purposes that prevented operation beyond the predesigned limits. To overcome this, we propose an omnidirectional mobile platform that can be connected to another platform. This type of modular structure can improve the efficiency of the entire logistics network by allowing flexible deployment according to the workload. The proposed system is able to generate holonomic movement with four mecanum wheels, and can connect with different mobile platforms using a connecting module at the side of each platform. To derive the kinematic model of the connected mobile platform, we first derive the analytical kinematic relationship between the connected platform center and each wheel. Next, we conducted simulations using RecurDyn, a dynamic simulation program, for three different types of connected mobile platforms. Our simulation results show that the position and velocity trajectory derived from the kinematic model well follows the reference trajectory at different connection cases.

This paper suggests a deep learning-based algorithm for monitoring workers’ safety in a smart factory environment. With the growth of smart factories in industry, the need for an AMR (autonomous mobile robot) that self-drives in a production line is increasing. Although most AMRs are designed to actively prevent a collision, there should be another monitoring solution to double-check workers’ safety because not all machines are reliable. We use an RGB-D camera which provides both depth information and RGB color information and a semantic segmentation method to monitor workers’ safety. The semantic segmentation algorithm is called Mask R-CNN and is used to detect workers and moveable equipment including AMRs. Since Mask R-CNN can specify an object’s boundary in RGB images, we are able to determine an object’s position in 3D coordinates by using the camera’s depth information. We can monitor the workers’ safety by checking whether they are close to hazardous equipment. We experimented with an AMR and manufacturing equipment to verify our suggested algorithm.

The purpose of autonomous driving is to help a mobile robot move from a starting point to a target point without any collisions. A mobile robot requires global path planning and local path planning to generate a path, where the local path planning guides the robot along the global path while taking action to avoid obstacles. In this paper, end-to-end approach to obstacle avoidance for a mobile robot is proposed. The approach uses a depth data to produce a straight motion, left rotation, or right rotation. The only data that the learning model requires is the depth image and control commands from the human operator. The method does not perform any additional operations like feature extraction on the raw sensor input values, resulting in a relatively low computing burden and intuitive learning without needing to know the robot model. Experimental results show that the proposed method successfully avoids obstacles.

The design of optimal input for fault identification in electric ground vehicles (EGVs) that use four independent in-wheel motors and four-wheel steering is presented in this paper. As the number of motors and steering actuators increases beyond the number of controlled variables, an EGV becomes an overactuated system, which provides actuator redundancy and the possibility of fault-tolerant control in the case of faults that occur in vehicle elements, such as its sensors and actuators. To ensure the reliability of the EGV during driving, online fault identification is needed, and its performance is directly dependent on the input signals. The input can be designed using the control allocation method, which is one approach to manage actuator redundancy. The proposed control allocation maximizes the sensitivity of the system output to parameters related to the fault position, while the system output is simultaneously controlled to maintain stability and follow the desired vehicle motions, even when faults occur. Simulations using the commercial software CarSim are performed to show the effectiveness of the proposed optimal input design method for fault identification; the performance of the system is compared with the conventional white noise perturbation input with equal power.