World Robot Research Articles

Optimization of Machine Learning-Based Automatic Target Detection and Locking System on Robots

Background: In recent years, the world of robotics has made significant progress in improving the operational capabilities of robots through target detection and locking systems. These systems play a crucial role in improving the efficiency and effectiveness of critical applications such as defense, security, and industrial automation. However, the main challenge faced is the limitations of the existing system in adapting to unstable environmental conditions and dynamic changes in targets. Objective: This research aims to overcome these challenges by developing a more adaptive and responsive target detection and locking system by integrating two leading machine learning technologies: Convolutional Neural Networks (CNN) for target detection and Long Short-Term Memory (LSTM) for target tracking. Methods: This study uses a quantitative approach to evaluate the effectiveness of the integration of CNNs and LSTMs in target detection and locking systems. Results: The results of the study showed a detection accuracy rate of 95% and a locking accuracy of 90%. The system is proven to be able to adapt to changing operational conditions in real-time and provide consistent performance in a variety of complex and dynamic scenarios. Conclusion: The conclusion of this study is that the integration of CNN and LSTM technologies in target detection and locking systems in robots significantly improves the performance and efficiency of the system, enabling a wider and more complex application.

Advancements in Autonomous Mobile Robot: A Holistic Review of Obstacle Avoidance Methods

The emergence of AMRs has altered our perspective and relationship with automation. At the heart of this transition is navigation and obstacle avoidance, both of which are important needs for deploying AMRs in a variety of scenarios. This comprehensive review looks at the latest advances in navigation and collision avoidance for AMRs, including a wide range of modern techniques and methodologies, algorithms, and technologies that aim to improve functionality. The study provides a detailed analysis of known approaches, such as rule-based approaches, potential fields, reactive navigation systems as behavior systems, and path-following algorithms, that have been developed to address the difficulty in practice. In contrast, technological advancements in machine learning, computer vision sensor fusion, and SLAM techniques, as well as edge computing, are reviewed in light of their unprecedented impact on AMR navigation. Global and local techniques are tackled using universal worldwide optics as well as national adaptations that reveal the unique characteristics of individual countries. The Data Analysis and Processing section emphasizes the importance of technologies that define AMR performance. Due to the constraints imposed by previous studies, it is clear that additional research is required to focus on closing gaps in controlled environments and using standard benchmarks; sensor heterogeneity issues; and practical implementation of theoretical aspects. In a nutshell, this review provides a map of the complex world of AMR navigation and obstacle avoidance. Its primary purpose is to contribute to the continuing debate, promote innovation, and suggest new research avenues in a fast-changing world of autonomous mobile robotics that breaks down traditional deployment constraints.

CRE-TSCAE: A Novel Classification Model Based on Stacked Convolutional Autoencoder for Dual-Target RSVP-BCI Tasks.

The Rapid Serial Visual Presentation (RSVP) paradigm facilitates target identification in a rapid picture stream, which is applied extensively in military target surveillance and police monitoring. Most researchers concentrate on the single target RSVP-BCI whereas the study of dual-target is scarcely conducted, limiting RSVP application considerably. This paper proposed a novel classification model named Common Representation Extraction-Targeted Stacked Convolutional Autoencoder (CRE-TSCAE) to detect two targets with one nontarget in RSVP tasks. CRE generated a common representation for each target class to reduce variability from different trials of the same class and distinguish the difference between two targets better. TSCAE aimed to control uncertainty in the training process while requiring less target training data. The model learned a compact and discriminative feature through the training from several learning tasks so as to distinguish each class effectively. It was validated on the World Robot Contest 2021 and 2022 ERP datasets. Experimental results showed that CRE-TSCAE outperformed the state-of-the-art RSVP decoding algorithms and the Average ACC was 71.25%, improving 6.5% at least over the rest. It demonstrated that CRE-TSCAE showed a strong ability to extract discriminative latent features in detecting the differences among two targets with nontarget, which guaranteed increased classification accuracy. CRE-TSCAE provided an innovative and effective classification model for dual-target RSVP-BCI tasks and some insights into the neurophysiological distinction between different targets.

2022 개정 고등학교 공학 선택과목에 대한 기술교사 교육요구도 분석과 지원 방안

Purpose: The purpose of this study was to analyze the level of technology teachers’ education needs for elective high school engineering subjects in the 2022 revised technology curriculum and derive specific demands for improving the expertise of technology teachers. Methods: In order to achieve the purpose of the study, literature related to the 2022 revised curriculum and education needs analysis was reviewed. A questionnaire was developed by the research team basedon the literature review and was revised and supplemented through a review by experts. The survey subjects of this study were limited to those eligible for the 2023 level 1 technology teacher qualification training. The survey tool asked technology teachers to respond to a 5-point Likert scale regarding the importance and feasibility of performance based on achievement standards for each subject. Educational needs were defined as the difference between the importance and performance scores. The IPA matrix method was used to determine the priority of support measures based on needs analysis. Results: The results obtained through this study are as follows. First, the demand for education in the world of robotics and engineering was the highest. Second, in the case of general intellectual property subjects, there was a high demand for education in specialized content area related to intellectual property rights. Third. In the case of the creative engineering design subject, there was a high demand for developing skills related to research and development projects in the engineering field. Conclusion: Through this study, it was determined that overall learning support is required for roboticsand the world of engineering, along with intensive training in intellectual property rights in the general intellectual property course, and project-based product development and prototyping in the creative engineering design course.0

Memorized Rapidly Exploring Random Tree Optimization (MRRTO): An Enhanced Algorithm for Robot Path Planning

Abstract With the advancement of the robotics world, many path-planning algorithms have been proposed. One of the important algorithms is the Rapidly Exploring Random Tree (RRT) but with the drawback of not guaranteeing the optimal path. This paper solves this problem by proposing a Memorized RRT Optimization Algorithm (MRRTO Algorithm) using memory as an optimization step. The algorithm obtains a single path from the start point, and another from the target point to store only the last visited new node. The method for computing the nearest node depends on the position, when a new node is added, the RRT function checks if there is another node closer to the new node rather than that is closer to the goal point. Simulation results with different environments show that the MRRTO outperforms the original RRT Algorithm, graph algorithms, and metaheuristic algorithms in terms of reducing time consumption, path length, and number of nodes used.

A ROS-Based Open Tool for Controlling an Educational Mobile Robot

Commercial educational robots provide an accessible entry point into the world of robotics. However, their programming is often limited to specific platforms, which can make it challenging to acquire the skills necessary for industry and research. In this study, we introduce an open-access tool developed using C++ and Arduino IDE that enables us to manage a commercial mobile robot through the Robot Operating System (ROS) middleware. This provides programmers with the ability to work in a powerful programming environment, such as Python. The robot used is the CrowBot BOLT, a kit based on ESP32 that enables wireless communication and includes various peripherals for application development. The mobile robot topics include robot velocities, RGB LEDs, a buzzer, a programmable button, and proximity, light, and line sensors. The proposal is assessed using two controllers: one for proximity and the other for tracking angular light. Both controllers are developed using Visual Studio Code. The experimental results demonstrated the proper functioning of the tool. Additionally, the response time was evaluated, and it was found that optimal performance is achieved at a frequency of 10 Hz. In summary, this proposal provides an accessible option for students and developers seeking to gain skills in robotics using ROS. The project’s repository is located at https://github.com/joseVarelaAldas/ROS-Crowbot.

Speed of Catch-Up and Convergence of the Artificial Intelligence Divide: AI Investment, Robotic, Start-Ups, and Patents

ABSTRACT With the recent acceleration in AI adoption and significant investments being made by many countries, the concern regarding the AI divide is becoming increasingly prevalent. We attempt to determine whether countries that have fallen behind in AI development can catch up to those leading the charge over time. Specifically, this study examines the dynamics of the AI divide on AI investments, robotics, AI start-ups, and AI patents in 34–57 countries, analyzing the data sets from Stanford AI Index Report, World Robotics Report, and the OECD Statistics website. We have used the methods of the relative AI divide (RAID) and absolute AI divide (AAID) for the two categories of income and adoption levels. Also, we have tested both γ convergence for the AI divide and σ convergence for the divide over time. We find that for all four AI categories, the relative AI divide between the leading and lagging countries has narrowed rapidly and the divide between the high- and middle-income subgroups has also narrowed at a slower pace. In terms of the micro convergence analysis of the AI divides for individual countries within the respective income and adoption subgroups, a change in the active ranking among countries was evident in nearly all of the subgroups. This study contributes to the literature by presenting not only the absolute AI divide but also the relative AI divide clarifying the changing trends systematically.

M-FANet: Multi-Feature Attention Convolutional Neural Network for Motor Imagery Decoding.

Motor imagery (MI) decoding methods are pivotal in advancing rehabilitation and motor control research. Effective extraction of spectral-spatial-temporal features is crucial for MI decoding from limited and low signal-to-noise ratio electroencephalogram (EEG) signal samples based on brain-computer interface (BCI). In this paper, we propose a lightweight Multi-Feature Attention Neural Network (M-FANet) for feature extraction and selection of multi-feature data. M-FANet employs several unique attention modules to eliminate redundant information in the frequency domain, enhance local spatial feature extraction and calibrate feature maps. We introduce a training method called Regularized Dropout (R-Drop) to address training-inference inconsistency caused by dropout and improve the model's generalization capability. We conduct extensive experiments on the BCI Competition IV 2a (BCIC-IV-2a) dataset and the 2019 World robot conference contest-BCI Robot Contest MI (WBCIC-MI) dataset. M-FANet achieves superior performance compared to state-of-the-art MI decoding methods, with 79.28% 4-class classification accuracy (kappa: 0.7259) on the BCIC-IV-2a dataset and 77.86% 3-class classification accuracy (kappa: 0.6650) on the WBCIC-MI dataset. The application of multi-feature attention modules and R-Drop in our lightweight model significantly enhances its performance, validated through comprehensive ablation experiments and visualizations.

Development of a remotely controllable 4 m long aerial-hose-type firefighting robot.

In a fire outbreak, firefighters are expected to rapidly extinguish fires to stop the spread of damage and prevent secondary disasters. We proposed the concept of a dragon firefighter (DFF), which is a flying-hose-type firefighting robot. We developed a 3.6 m long DFF equipped with two nozzle units and achieved stable flight. However, the system was not yet completed because the root of the robot, which should have been operated remotely, was operated manually. In addition, the system's reliability was insufficient to successfully repeat the demonstration several times. The development of a robot demonstration system is crucial for the practical application of such a firefighting robot. In this study, we developed a demonstration system for a remotely controllable 4 m flying firehose robot for demonstration at the World Robot Summit 2020 (WRS 2020) opening ceremony in Fukushima as a milestone. This paper focuses on the following issues: 1): installation of the remotely controllable mobile base, 2): redesign of the water channels (the sizes of nozzle outlets) to get enough thrusts to fly with a fire engine, 3): development of nozzle units with a larger movable range (1.5 times larger than the conventional nozzle) in addition to waterproofing technique to improve system reliability, and 4): redesign of a passive damping mechanism to ensure better stability. Thus, a firefighting demonstration was successfully conducted at the opening ceremony of the World Robot Summit 2020 in Fukushima, Japan, and we discuss the lessons learned through the demonstration. We found that the developed DFF system incorporating a mobile base could achieve remote fire extinguishing.

PELATIHAN OBSTACLE AVOIDER ROBOT UNTUK PENGENALAN DUNIA ROBOTIKA BAGI SISWA SMAN 2 KABUPATEN TANGERANG

Integrating robotics into the education curriculum is a progressive step that can open doors for academic and student progress. The problems faced by extracurricular students at KIR SMAN 2 Tangerang Regency related to integrating robotics with academics have yet to be achieved. To facilitate students' academic and professional progress, the solution to this problem is to hold robotics training to introduce the world of robotics. The methods used include presentations, discussions, demonstrations, and practicums. The results of this research produced students who were able to assemble an Arduino obstacle avoider robot.

A Q-learning approach to the continuous control problem of robot inverted pendulum balancing

This study evaluates the application of a discrete action space reinforcement learning method (Q-learning) to the continuous control problem of robot inverted pendulum balancing. To speed up the learning process and to overcome technical difficulties related to the direct learning on the real robotic system, the learning phase is performed in simulation environment. A mathematical model of the system dynamics is implemented, deduced by curve fitting on data acquired from the real system. The proposed approach demonstrated feasible, featuring its application on a real world robot that learned to balance an inverted pendulum. This study also reinforces and demonstrates the importance of an accurate representation of the physical world in simulation to achieve a more efficient implementation of reinforcement learning algorithms in real world, even when using a discrete action space algorithm to control a continuous action.

Evolutionary swarm formation: From simulations to real world robots

Swarms of autonomous robots have become an interesting alternative for space and aerospace applications due to their versatility, robustness, and self-organising capability. Some of those applications, such as asteroid observation, convoy escort, and counter-drone systems, rely on stable formations achieved around a central point of interest. However, the use of different numbers of robots and the existence of a wide range of initial conditions contribute to make it a challenging problem. We propose in this research work a novel approach for self-organising a swarm of autonomous robots where the members’ movements depend only on their relative position (range and bearing) obtained from their respective radio beacons. An optimisation approach based on an evolutionary algorithm is proposed to calculate the optimal swarm’s parameters, e.g. speed and attracting/repelling forces, to achieve robust formations under different initial conditions and failure rates. Experiments are conducted using realistic simulations of six case studies featuring three, five, ten, fifteen, twenty, and thirty robots. The best valued configurations were tested on 420 scenarios showing that our proposal is robust since it has always achieved the desired circular formation. Finally, we have used real E-Puck2 robots to validate the swarm’s capability of self-organising around a central point of interest as well as its resilience to robot failure, obtaining successful circular formations in all the experiments.

Robotic world models-conceptualization, review, and engineering best practices.

The term "world model" (WM) has surfaced several times in robotics, for instance, in the context of mobile manipulation, navigation and mapping, and deep reinforcement learning. Despite its frequent use, the term does not appear to have a concise definition that is consistently used across domains and research fields. In this review article, we bootstrap a terminology for WMs, describe important design dimensions found in robotic WMs, and use them to analyze the literature on WMs in robotics, which spans four decades. Throughout, we motivate the need for WMs by using principles from software engineering, including "Design for use," "Do not repeat yourself," and "Low coupling, high cohesion." Concrete design guidelines are proposed for the future development and implementation of WMs. Finally, we highlight similarities and differences between the use of the term "world model" in robotic mobile manipulation and deep reinforcement learning.

Analysis of the performance of robotic assembly systems considering the usage of tools or jigs for the task-board task in the WRS 2018

ABSTRACT A task-board task is a task in which robots perform a set of fundamental assembly operations related to the assembly of a specific product such as inserting a part into another part, tightening bolts and nuts, attaching flexible parts, etc. In the task-board task competition in the industrial robotics category of the World Robot Summit (WRS) 2018, 15 teams from around the world performed the task. In our previous work, we observed recorded videos of the competition frame by frame and manually investigated the performance of robotic systems of the top four teams, focusing only on whether robots contact target parts. In this paper, we introduce frame sets to represent mainly contact states between robots, tools, or jigs and target parts to be handled for such a frame-by-frame analysis of the recorded videos, and based on the frame sets, we analyze the performance of robotic assembly systems of other teams as well as the top four teams in the task-board task competition in the WRS 2018, considering the usage of tools or jigs.

An approach to forecasting promising areas of development of the robotics industry based on statistical reports of the International Federation of Robotics (IFR) in conditions of economic uncertainty

A simple method of forming a forecast rating of the most promising directions of robotics development at the mo-ment is proposed, depending on the current market situation affected by unstable economic factors. The ad-vantage of the considered methodology is that, firstly, a relevant-processed array of data regularly published in the annual statistical collections of IFR - World Robotics is used as the basis of the analysis, secondly, the availability of varying preferences of factors affecting the economic situation, and, thirdly– simplicity. The results obtained can be used by interested organizations both for the development of conceptual documents on the development of robot-ics, and for the development of proposals for the initiation of commercial projects.

A review of deep learning methods for cross-subject rapid serial visual presentation detection in World Robot Contest 2022

The rapid serial visual presentation (RSVP) paradigm has garnered considerable attention in brain–computer interface (BCI) systems. Studies have focused on using cross-subject electroencephalogram data to train cross-subject RSVP detection models. In this study, we performed a comparative analysis of the top 5 deep learning algorithms used by various teams in the event-related potential competition of the BCI Controlled Robot Contest in World Robot Contest 2022. We evaluated these algorithms on the final data set and compared their performance in cross-subject RSVP detection. The results revealed that deep learning models can achieve excellent results with appropriate training methods when applied to cross-subject detection tasks. We discussed the limitations of existing deep learning algorithms in cross-subject RSVP detection and highlighted potential research directions.

Overview of recognition methods for SSVEP-based BCIs in World Robot Contest 2022: MATLAB undergraduate group

The steady-state visual evoked potential (SSVEP)-based speller has emerged as a widely adopted paradigm in current brain–computer interface (BCI) systems due to its rapid processing and consistent performance across different individuals. Calibration-free SSVEP algorithms, as opposed to their calibration-based counterparts, offer clear and intuitive mathematical principles, making them accessible to novice developers. During the World Robot Contest (WRC) 2022, participants in the undergraduate category utilized various approaches to accomplish target detection in the calibration-free setting, successfully implementing the algorithms using MATLAB. The winning approach achieved an average information transfer rate of 198.94 bits/min in the final test, which is notably high given the calibration-free scenario. This paper presents an introduction to the underlying principles of the selected methods, accompanied by a comparison of their effectiveness through analysis of results from both the final test and offline experiments. Additionally, we propose that the youth competition of WRC could serve as an ideal starting point for beginners interested in studying and developing their own BCI systems.

Winning algorithms in BCI Controlled Robot Contest in World Robot Contest 2022: BCI Turing Test

The Turing Test is a method of testing whether a machine has human intelligence. A novel brain–computer interface (BCI) Turing Test was proposed in the BCI Controlled Robot Contest in World Robot Contest 2022. Contestants developed algorithms that can distinguish if an instruction is issued by a human. Participants collaborated with an electroencephalogram-based BCI to play a soccer game in a virtual scenario. Participants were asked to perform steady-state visual evoked potential (SSVEP) tasks or motor imagery (MI) tasks to control the robots or be in an idle state to mimic the system giving instructions on behalf of the participants. Several algorithms proposed in this competition are developed based on the concept that the idle state is a category in multiclass classification problems. This paper details the algorithms of the top five teams with the best performance in the final, lists the popular classification models and algorithms for MI and SSVEP, and discusses the effectiveness of each approach in improving classification performance and reducing the computation time.

Algorithm contest of motor imagery BCI in the World Robot Contest 2022: A survey

From August 19 to 21, 2022, the BCI Controlled Robot Contest finals in the World Robot Contest 2022 were held in Beijing, China. Fifteen teams participated in the finals in the Algorithm Contest of Motor Imagery BCI. This paper introduces the algorithms in the motor imagery (MI) classification area, describes the competition content and set, and summarizes the algorithms and results of the top five teams in the finals. First, the MI paradigm and the overview of the existing motor imagery brain–computer interface classification algorithms are introduced, followed by the introduction of the algorithms of the top five teams in the final step by step, including electroencephalography channel selection, data length selection, data preprocessing, data augmentation, classification network, training, and testing settings. Finally, the highlights and results of each algorithm are discussed.

Overview of the winning approaches in 2022 World Robot Contest Championship–Asynchronous SSVEP

In recent years, the steady-state visual evoked potential (SSVEP) electroencephalogram paradigm has gained considerable attention owing to its high information transfer rate. Several approaches have been proposed to improve the performance of SSVEP-based brain–computer interface (BCI) systems. In SSVEP-based BCIs, the asynchronous scenario poses a challenge as the subjects stare at the screen without synchronization signals from the system. The algorithm must distinguish whether the subject is being stimulated or not, which presents a significant challenge for accurate classification. In the 2022 World Robot Contest Championship, several effective algorithm frameworks were proposed by participating teams to address this issue in the SSVEP competition. The efficacy of the approaches employed by five teams in the final round is demonstrated in this study, and an overview of their methods is provided. Based on the final score, this paper presents a comparative analysis of five algorithms that propose distinct asynchronous recognition frameworks via diverse statistical methods to differentiate between intentional control state and non-control state based on dynamic window strategies. These algorithms achieve an impressive information transfer rate of 89.833 and a low false positive rate of 0.073. This study provides an overview of the algorithms employed by different teams to address asynchronous scenarios in SSVEP-based BCIs and identifies potential future avenues for research in this area.