KeJia is a domestic service robot, consisting of a mobile base, an arm, two cameras, and a set of software components for perception, manipulation, natural language understanding, motion and task planning, and decision making. With on-line running of these functions, a robot can adapt to dynamic environments which may have unexpected changes. In this paper, we propose a novel hierarchical method which combines motion planning with a neural network, so that the robot can tolerate errors from sensors, wear of parts, and human disturbances during motion execution. We evaluate our work on KeJia that cooks popcorn using a microwave oven, where humans try to disturb KeJia during the operation.
Today, exoskeletons are widely applied to provide walking assistance for patients with lower limb motor incapacity. Most existing exoskeletons are under-actuated, resulting in a series of problems, e.g., interference and unnatural gait during walking. In this study, we propose a novel intelligent autonomous lower extremity exoskeleton (Auto-LEE), aiming at improving the user experience of wearable walking aids and extending their application range. Unlike traditional exoskeletons, Auto-LEE has 10 degrees of freedom, and all the joints are actuated independently by direct current motors, which allows the robot to maintain balance in aiding walking without extra support. The new exoskeleton is designed and developed with a modular structure concept and multi-modal human-robot interfaces are considered in the control system. To validate the ability of self-balancing bipedal walking, three general algorithms for generating walking patterns are researched, and a preliminary experiment is implemented.
There is an ocean current in the actual underwater working environment. An improved self-organizing neural network task allocation model of multiple autonomous underwater vehicles (AUVs) is proposed for a three-dimensional underwater workspace in the ocean current. Each AUV in the model will be competed, and the shortest path under an ocean current and different azimuths will be selected for task assignment and path planning while guaranteeing the least total consumption. First, the initial position and orientation of each AUV are determined. The velocity and azimuths of the constant ocean current are determined. Then the AUV task assignment problem in the constant ocean current environment is considered. The AUV that has the shortest path is selected for task assignment and path planning. Finally, to prove the effectiveness of the proposed method, simulation results are given.
Recognizing and predicting the movement and intention of the wearer in control of an exoskeleton robot is very challenging. It is difficult for exoskeleton robots, which measure and drive human movements, to interact with humans. Therefore, many different types of sensors are needed. When using various sensors, a data design is needed for effective sensing. An electromyographic (EMG) signal can be used to identify intended motion before the actual movement, and the delay time can be shortened via control of the exoskeleton robot. Before using a lower limb exoskeleton to help in walking, the aim of this work is to distinguish the walking environment and gait period using various sensors, including the surface electromyography (sEMG) sensor. For this purpose, a gait experiment was performed on four subjects using the ground reaction force, human–robot interaction force, and position sensors with sEMG sensors. The purpose of this paper is to show progress with the use of sEMG when recognizing walking environments and the gait period with other sensors. For effective data design, we used a combination of sensor types, sEMG sensor locations, and sEMG features. The results obtained using an individual mechanical sensor together with sEMG showed improvement compared to the case of using an individual sensor, and the combination of sEMG and position information showed the best performance in the same number of combinations of three sensors. When four sensor combinations were used, the environment classification accuracy was 96.1%, and the gait period classification accuracy was 97.8%. Vastus medialis (VM) and gastrocnemius (GAS) were the most effective combinations of two muscle types among the five sEMG sensor locations on the legs, and the results were 74.4% in pre-heel contact (preHC) and 71.7% in pre-toe-off (preTO) for environment classification, and 68.0% for gait period classification, when using only the sEMG sensor. The two effective sEMG feature combinations were “mean absolute value (MAV), zero crossings (ZC)” and “MAV, waveform length (WL)”, and the “MAV, ZC” results were 80.0%, 77.1%, and 75.5%. These results suggest that the sEMG signal can be effectively used to control an exoskeleton robot.
A new hierarchical software architecture is proposed to improve the safety and reliability of a safetycritical drone system from the perspective of its source code. The proposed architecture uses formal verification methods to ensure that the implementation of each module satisfies its expected design specification, so that it prevents a drone from crashing due to unexpected software failures. This study builds on top of a formally verified operating system kernel, certified kit operating system (CertiKOS). Since device drivers are considered the most important parts affecting the safety of the drone system, we focus mainly on verifying bus drivers such as the serial peripheral interface and the inter-integrated circuit drivers in a drone system using a rigorous formal verification method. Experiments have been carried out to demonstrate the improvement in reliability in case of device anomalies.
Robots need task planning algorithms to sequence actions toward accomplishing goals that are impossible through individual actions. Off-the-shelf task planners can be used by intelligent robotics practitioners to solve a variety of planning problems. However, many different planners exist, each with different strengths and weaknesses, and there are no general rules for which planner would be best to apply to a given problem. In this study, we empirically compare the performance of state-of-the-art planners that use either the planning domain description language (PDDL) or answer set programming (ASP) as the underlying action language. PDDL is designed for task planning, and PDDL-based planners are widely used for a variety of planning problems. ASP is designed for knowledge-intensive reasoning, but can also be used to solve task planning problems. Given domain encodings that are as similar as possible, we find that PDDL-based planners perform better on problems with longer solutions, and ASP-based planners are better on tasks with a large number of objects or tasks in which complex reasoning is required to reason about action preconditions and effects. The resulting analysis can inform selection among general-purpose planning systems for particular robot task planning domains.
There is an increasing need to introduce socially interactive robots as a means of assistance in autism spectrum disorder (ASD) treatment and rehabilitation, to improve the effectiveness of rehabilitation training and the diversification of treatment, and to alleviate the shortage of medical personnel in mainland China and other places in the world. In this preliminary clinical study, three different socially interactive robots with different appearances and functionalities were tested in therapy-like settings in four different rehabilitation facilities/institutions in Shenzhen, China. Seventy-four participants, including 52 children with ASD, whose processes of interacting with robots were recorded by three different cameras, all received a single-session three-robot intervention. Data were collected from not only the videos recorded, but also the questionnaires filled mostly by parents of the participants. Some insights from the preliminary results were obtained. These can contribute to the research on physical robot design and evaluations on robots in therapy-like settings. First, when doing physical robot design, some preferential focus should be on aspects of appearances and functionalities. Second, attention analysis using algorithms such as estimation of the directions of gaze and head posture of a child in the video clips can be adopted to quantitatively measure the prosocial behaviors and actions (e.g., attention shifting from one particular robot to other robots) of the children. Third, observing and calculating the frequency of the time children spend on exploring/playing with the robots in the video clips can be adopted to qualitatively analyze such behaviors and actions. Limitations of the present study are also presented.
Recently, sequence anomaly detection has been widely used in many fields. Sequence data in these fields are usually multi-dimensional over the data stream. It is a challenge to design an anomaly detection method for a multi-dimensional sequence over the data stream to satisfy the requirements of accuracy and high speed. It is because: (1) Redundant dimensions in sequence data and large state space lead to a poor ability for sequence modeling; (2) Anomaly detection cannot adapt to the high-speed nature of the data stream, especially when concept drift occurs, and it will reduce the detection rate. On one hand, most existing methods of sequence anomaly detection focus on the single-dimension sequence. On the other hand, some studies concerning multi-dimensional sequence concentrate mainly on the static database rather than the data stream. To improve the performance of anomaly detection for a multi-dimensional sequence over the data stream, we propose a novel unsupervised fast and accurate anomaly detection (FAAD) method which includes three algorithms. First, a method called “information calculation and minimum spanning tree cluster” is adopted to reduce redundant dimensions. Second, to speed up model construction and ensure the detection rate for the sequence over the data stream, we propose a method called “random sampling and subsequence partitioning based on the index probabilistic suffix tree.” Last, the method called “anomaly buffer based on model dynamic adjustment” dramatically reduces the effects of concept drift in the data stream. FAAD is implemented on the streaming platform Storm to detect multi-dimensional log audit data. Compared with the existing anomaly detection methods, FAAD has a good performance in detection rate and speed without being affected by concept drift.
Reconstruction of a 12-lead electrocardiogram (ECG) from a serial 3-lead ECG has been researched in the past to satisfy the need for more wearing comfort and ambulatory situations. The accuracy and real-time performance of traditional methods need to be improved. In this study, we present a novel method based on convolutional neural networks (CNNs) for the synthesis of missing precordial leads. The results show that the proposed method receives better similarity and consumes less time using the PTB database. Particularly, the presented method shows outstanding performance in reconstructing the pathological ECG signal, which is crucial for cardiac diagnosis. Our CNN-based method is shown to be more accurate and time-saving for deployment in non-hospital situations to synthesize a standard 12-lead ECG from a reduced lead-set ECG recording. This is promising for real cardiac care.
A novel topology Halbach permanent magnet array is proposed and applied to the design of a printed circuit board (PCB) axial flux permanent magnet (AFPM) motor. Compared with the traditional coreless AFPM motor, this novel topology for a Halbach permanent magnet array PCB stator AFPM motor has larger air-gap magnetic flux density and air-gap flux per pole. The magnetic flux leakage is effectively reduced, and the air-gap magnetic density is close to the sine wave. Results of the finite element analysis and prototype experiments verify the feasibility and effectiveness of the novel Halbach permanent magnet array PCB stator motor. A reference basis and practical value for the design of the PCB AFPM motor are provided.
The threats and challenges of unmanned aerial vehicle (UAV) invasion defense due to rapid UAV development have attracted increased attention recently. One of the important UAV invasion defense methods is radar network detection. To form a tight and reliable radar surveillance network with limited resources, it is essential to investigate optimized radar network deployment. This optimization problem is difficult to solve due to its nonlinear features and strong coupling of multiple constraints. To address these issues, we propose an improved firefly algorithm that employs a neighborhood learning strategy with a feedback mechanism and chaotic local search by elite fireflies to obtain a trade-off between exploration and exploitation abilities. Moreover, a chaotic sequence is used to generate initial firefly positions to improve population diversity. Experiments have been conducted on 12 famous benchmark functions and in a classical radar deployment scenario. Results indicate that our approach achieves much better performance than the classical firefly algorithm (FA) and four recently proposed FA variants.