Springer Science and Business Media LLC

Although it has been well known that novices should train a good lifting posture, there was little way to recognize whether the current posture was good or not based on measured data. The purpose of this paper was to classify the difference between skilled experts working at a freight transport company and unskilled novices without any experience during symmetric lifting by using center-of-pressure (CoP) velocities. All the human subjects performed symmetric lifting experiments with closed eyes; the experiments involved lifting loads (6 and 18 kg) to the upside. Time series data of the CoP position were measured, using a Wii Balance Board, and then, the CoP velocities were calculated. The linear discriminant analysis (LDA) was designed by seven indices which were derived from CoP velocities that reflected the center-of-mass acceleration. The result indicated that the designed LDA discriminated the difference in posture between the two groups with the low error rate (0.100 and 0.017) for classification under 6 and 18 kg. Based on measurement results of CoP trajectories, we inferred that the difference in the CoP velocities between the two groups could be attributed to the difference in the balance ability which means that most skilled experts place their body weight on their rearfeet during symmetric lifting. The LDA classifier designed by CoP velocities was helpful for recognition of the difference between skilled experts and unskilled novices during symmetric lifting. Because the skillful characteristics of experts may be responsible for the lightening of the burden on the waist during lifting, it is considered for the regular check of posture to be helpful for reducing the ratio of occupational low back pain at the workplace.

Seasonal flu is currently a major public health issue the world is facing. Although the World Health Organization (WHO) suggests social distancing is one of the best ways to stop the spread of the flu disease, the lack of controllability in keeping a social distance is widespread. Spurred by this concern, this paper developed a fast social distancing monitoring solution, which combines a lightweight PyTorch-based monocular vision detection model with inverse perspective mapping (IPM) technology, enabling the nursing robot to recover 3D indoor information from a monocular image and detect the distance between pedestrians, then conducts a live and dynamic infection risk assessment by statistically analyzing the distance between the people within a scene and ranking public places into different risk levels, called Fast DeepSOCIAL (FDS). First, the FDS model generates the probability of an object’s category and location directly using a lightweight PyTorch-based one-stage detector, which enables a nursing robot to obtain significant real-time performance gains while reducing memory consumption. Additionally, the FDS model utilizes an improved spatial pyramid pooling strategy, which introduces more branches and parallel pooling with different kernel sizes, which will be beneficial in capturing the contextual information at multiple scales and thus improving detection accuracy. Finally, the nursing robot introduces a gap-seeking strategy based on obstacles-weighted control (GSOWC) to adapt to dangerous indoor disinfection tasks while quickly avoiding obstacles in an unknown and cluttered environment. The performance of the FDS on the nursing robot platform is verified through extensive evaluation, demonstrating its superior performance compared to seven state-of-the-art methods and revealing that the FDS model can better detect social distance. Overall, a nursing robot employing the Fast DeepSOCIAL model (FDS) will be an innovative approach that effectively contributes to dealing with this seasonal flu disaster due to its fast, contactless, and inexpensive features.

This paper addresses the design of lightweight radiation sensors for the small-scale unmanned aerial system (UAS) and its implementation for low-altitude radiation source localization and contour mapping. The compact high-resolution gamma-ray CZT sensors were integrated into UAS platforms as plug-and-play components using robot operating system. The swarm of UAS has advantages over a single agent-based approach in detecting radiative sources and effectively mapping the area. The proposed swarm consists of three UAS platforms in a circular formation. The proposed approach can potentially be used for low-altitude clustered environments where a conventional helicopter-based platform cannot be utilized. It can provide a relatively precise boundary of the safe area for potential human exploration as well as enhancing situation awareness capabilities for first responders. The source seeking and contour mapping algorithms are developed based on a simple 1/R2 radiation field, but they are validated in more realistic radiation field having multiple sources and physical structures with scattering and attenuation effects simulated by MCNP code. Also, gradient estimation and contour mapping algorithms are validated experimentally with small-scale multicopter platforms in the indoor flight testbed.

This paper presents an approach to optimize the control torque of heavy-duty redundant manipulators used for dismantling nuclear power plants. Such manipulators must endure intensive and repetitive tasks over long periods. In this regard, the torque minimization is essential for decreasing power consumption and the fatigue load acting on the joint bearings. This in turn can increase the lifespan of the manipulators and lead to saving on maintenance costs. Because of the design specifications of the manipulators, gravity entirely dominates the Coriolis and centrifugal torques. Hence, it is challenging to reduce the driving torque through the application of a conventional optimization method, known as the minimum kinetic energy method, where the configuration of the manipulator changes extremely slowly from the beginning to the end of the trajectory. In this study, we propose a new torque minimization method based on the advantage of the redundancy of the manipulator. In particular, the norm of the static torque caused by the manipulator gravity itself tends to decrease owing to the application of the gradient projection method for the redundancy resolution at the acceleration level. Simultaneously, the dynamic torque is minimized to lessen the local acceleration triggered by the change mentioned above. The generalized effectiveness of this proposed method is evaluated through simulations and experiments with two different trajectories and speeds. The results show that the proposed method is more effective in reducing the overall driving torque and dissipated energy compared with the conventional technique, especially in the case of the 7-DOF heavy-duty redundant manipulator, and would be applicable for the revolute robot type.

The recent increase in technological maturity has empowered robots to assist humans and provide daily services. Voice command usually appears as a popular human–machine interface for communication. Unfortunately, deaf people cannot exchange information from robots through vocal modalities. To interact with deaf people effectively and intuitively, it is desired that robots, especially humanoids, have manual communication skills, such as performing sign languages. Without ad hoc programming to generate a particular sign language motion, we present an imitation system to teach the humanoid robot performing sign languages by directly replicating observed demonstration. The system symbolically encodes the information of human hand–arm motion from low-cost depth sensors as a skeleton motion time-series that serves to generate initial robot movement by means of perception-to-action mapping. To tackle the body correspondence problem, the virtual impedance control approach is adopted to smoothly follow the initial movement, while preventing potential risks due to the difference in the physical properties between the human and the robot, such as joint limit and self-collision. In addition, the integration of the leg-joints stabilizer provides better balance of the whole robot. Finally, our developed humanoid robot, NINO, successfully learned by imitation from human demonstration to introduce itself using Taiwanese Sign Language.

This paper presents a novel impedance control strategy to improve the performance of a robot manipulator. Impedance control and admittance control have complementary effects on the stability and performance of a control system. Impedance control works well in stiff environments, whereas admittance control works well in soft environments. In this paper, we propose a hybrid impedance and admittance control strategy that switches the controller based on the switching condition. If proper switching between impedance control and admittance control is achieved, the controller will have the advantages of both the control strategies. The proposed schemes were evaluated through simulations using a 2-DOF manipulator. Experiments were conducted using an actual robot. The results of the simulation and experiments performed confirmed that the proposed control strategy improves the performance of the robot manipulator.

In this paper, we propose a new localization algorithm based on a hybrid trilateration algorithm for obtaining an accurate position of a robot in intelligent space. The proposed algorithm is also able to estimate a position of the moving robot by using the extended Kalman filter, taking into consideration time synchronization and velocity of the robot. For realizing the localization system, we employ several smart sensors as beacons on the ceiling in intelligent space and as a listener attached to the robot. Finally, simulation results show the feasibility and effectiveness of the proposed localization algorithm compared with existing trilateration algorithms.

The human body can be modeled as a kinematically redundant manipulator which exploits “redundant degrees of freedom” to execute various motions in a suitable fashion. Differently from the typical kinematically redundant robots that are attached to the fixed ground, the zero moment point (ZMP) condition should be taken into account not to fall down. Thus, this paper investigates a motion planning algorithm for kinematically redundant manipulator standing on the ground. For this, a geometric constraint equation is derived from the existing ZMP equation. This constraint equation is formed like a second-order kinematic equation, which enables one to plan the ZMP trajectory in a feed-forward fashion. This constraint equation and the kinematic equation of the manipulator model are solved together. Then, the solution of this composite equation guarantees both the desired operational motion and the ZMP trajectory. The feasibility of the proposed algorithms is verified by simulating and experimenting several motions though a planar 3-DOF manipulator model.