The maritime industry's continuous commitment to sustainability has led to a dedicated exploration of methods to reduce vessel fuel consumption. This paper undertakes this challenge through a machine learning approach, leveraging a real-world dataset spanning two years of a passenger vessel in west coast Canada. Our focus centers on the creation of a time series forecasting model given the dynamic and static states, actions, and disturbances. This model is designed to predict dynamic states based on the actions provided, subsequently serving as an evaluative tool to assess the proficiency of the vessel's operation under the captain's guidance. Additionally, it lays the foundation for future optimization algorithms, providing valuable feedback on decision-making processes. To facilitate future studies, our code is available at https://github.com/pagand/model_optimze_vessel/tree/AAAI.

We propose a novel methodology for robotic follow-ahead applications that address the critical challenge of obstacle and occlusion avoidance. Our approach effectively navigates the robot while ensuring avoidance of collisions and occlusions caused by surrounding objects. To achieve this, we developed a high-level decision-making algorithm that generates short-term navigational goals for the mobile robot. Monte Carlo Tree Search is integrated with a Deep Reinforcement Learning method to enhance the performance of the decision-making process and generate more reliable navigational goals. Through extensive experimentation and analysis, we demonstrate the effectiveness and superiority of our proposed approach in comparison to the existing follow-ahead human-following robotic methods. Our code is available at https://github.com/saharLeisiazar/follow-ahead-ros.

Brain-computer interface (BCI) technology, particularly those based on electroencephalography (EEG), holds significant potential for controlling powered lower limb exoskeletons in rehabilitation contexts. This study introduces an EEG-based BCI system designed to decode anticipated gait direction, thereby enabling command of a self-balancing, overground exoskeleton. To evaluate the performance of the proposed system safely and effectively, we utilized a dynamic simulator (or a digital twin) of the physical exoskeleton, developed commercially for individuals with limited or no walking ability. Six healthy participants, wearing an EEG device, were instructed to initiate gait movements toward the direction indicated by on-screen arrow triggers (forward, backward, left, and right). A Convolutional Neural Network (CNN), operating on an 80%-20% Train-Test Ratio, was used to evaluate the system. The results demonstrated low error rates for the exoskeleton simulator, and an overall system accuracy of 0.75, reflecting the performance of the EEG-based BCI. Notably, the system had an average delay of 5 minutes in a real-time control setting, primarily attributed to its signal processing step.

The available human body inertial motion capture (MoCap) systems are aided by magnetometers to remove the drift error in yaw angle estimation, which in turn limits their application in the presence of long-lasting magnetic disturbances in indoor environments. This paper introduces a magnetometer-free algorithm for lower-body MoCap including 3-D localization and posture tracking by fusing inertial sensors with an ultrawideband (UWB) localization system and a biomechanical model of the human lower body. Using our novel Kalman filter-based fusion algorithm, the UWB localization data aided by the biomechanical model can eliminate the drift in inertial yaw angle estimation of the lower-body segments. This magnetometer-free yaw angle estimation makes the algorithm insensitive to the magnetic disturbances. The algorithm is benchmarked against the optical MoCap system for various indoor activities including walking, jogging, jumping, and stairs ascending/descending. The results show that the proposed algorithm outperforms the available magnetometer-aided algorithms in yaw angle tracking under magnetic disturbances. In a uniform magnetic field, the algorithm shows similar accuracies in localization and joint angle tracking when compared with the magnetometer-aided methods. The localization accuracy of the proposed method is better than 4.5 cm in a 3-D space and its accuracy for knee angle tracking is about 3.5° and 4.5° in low and high dynamic motions, respectively.

This work presents a fully automated method for detecting the contact between a microcapillary tip and a cell membrane based on a statistical process control (SPC) algorithm known as the double-sided cumulative sum (or "cusum"). By analyzing current measurements obtained through a microcapillary electrode, the proposed goal of this system is to determine when tip-to-membrane (tip-membrane) contact occurs using thin adhered cells (e.g., less than 10 μm) for the purposes of fully automated robotic-assisted, single cell electroporation (SCE) - a powerful method of gene transfection. This SPC algorithm is robust against uncontrollable system parameters such as system noise common in electrode-based systems, nonstationary processes, and variations in the physical parameters of cells. The proposed algorithm was successfully demonstrated on adhered mammal cells as small as 4 μm in thickness and using tip-placement velocities from 1 to 8 μm/s. In addition, a novel method of experimentation is described correlating optical measurements between tip-membrane proximity and changes in icctduring the tip-placement sequence.

In this paper, a computationally efficient orientation estimation algorithm using an inertial/magnetic sensor is presented for ambulatory real-time human motion tracking. Based on a quaternion formulation, the proposed algorithm is designed to have two main steps that are connected in feedback relationship: a quaternion measurement step with a vector selector scheme and a Kalman filter (KF) step. This allows us to choose only the quaternion as the state and measurement vectors in our KF design. Thus, the KF has a minimum-order structure (i.e., 4th-order), which decreases the computational cost. The estimated orientation accuracy is validated experimentally by using an optical tracking system.

This paper describes a new quaternion-based Kalman filter (KF) for estimating human body orientation using an inertial/magnetic sensor. The proposed algorithm is comprised of a quaternion measurement step and a KF step that are connected in feedback relationship. This allows the algorithm to have a minimum-order structure (i.e., fourth order) that is computationally very efficient. Furthermore, to offer more reliable information to the quaternion measurement step, a vector selector scheme is adopted, which effectively adds the gyro measurement to the so-called Wahba's problem that conventionally uses only the accelerometer and magnetometer measurements. This protects the algorithm against undesirable conditions such as fast movements and temporary magnetic disturbances, enabling it to compute an accurate orientation estimate. Due to the computational efficiency of the algorithm, it is suitable for real-time ambulatory human motion tracking applications that require multiple and untethered inertial/magnetic sensors with low-cost onboard processing.

Studies of individual cells via microscopy and microinjection are a key component in research on gene functions, cancer, stem cells, and reproductive technology. As biomedical experiments become more complex, there is an urgent need for robotic systems to: improve cell manipulation, increase throughput, reduce lost cells, and improve reaction detection. Automation of these tasks using visual servoing creates significant benefits for biomedical laboratories including repeatability of experiments, higher throughput, and a controlled environment capable of operating 24 hours a day. In this work the design and development of a new five degree-of- freedom biomanipulator designed for single-cell microinjection, is described. The biomanipulator employs three degrees of linear motion and two degrees of rotation. This provides the ability to manipulate/micro-inject cells at varying orientations, thereby increasing flexibility in dealing with complex operations such as injecting clustered cells. The capability of the biomanipulator is shown with preliminary experimental results using mouse myeloma cells.

This paper presents the development and demonstration of a robotic gripper for active control of shape and vibration of thin-walled flexible payloads. The gripper is a smart gripper, which is configured with multiple actuated fingers that are comprised of linear actuators with DC motors, and laser proximity sensors to enable control of both shape and vibration of thin-walled flexible payloads. A detailed description of the proof-of-concept smart gripper is presented, followed by an experimental demonstration using an automotive front quarter body panel.

This paper presents a theoretical framework to simultaneously control both static shape deformation and vibration of a flexible payload grasped by an actuated robotic gripper. The dynamic model of the actuated flexible payload is derived using the component mode synthesis (CMS) method with addition of quasi-static modes. Then, the dynamic component model is used to synthesize a simultaneous static shape and vibration controller. Two-time scale control scheme is pursued taking advantage of the two-time scale behavior between the quasi-static modes and vibration modes, which are employed in the model. Simulation results demonstrate the effectiveness of the proposed control approach to simultaneously correct static deformation and suppress vibration in the payload.

This paper presents an improved method of modeling the linear dynamics of a flexible payload grasped by an actuated robotic gripper. The component mode synthesis (CMS) method is employed to explicitly model the coupling between the payload and actuators, and to reduce the system order. With the addition of quasi-static modes to vibration normal modes and constraint or attachment modes, as pre-selected component modes, a new component mode representation is defined. In this study, it is found that the inclusion of quasi-static modes and attachment modes in the CMS formulation results in increased accuracy for simulation of dynamic responses of flexible payloads subject to both static and dynamic external forces. Numerical examples are presented to demonstrate the effectiveness of the new component mode representation for the given problem.