Robot Contest Research Articles

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.

Four Wheel Omni Trajectory with Gyrodometry Method in the Indonesia 2022 Ash Robot Contest

Four Wheel Omni Trajectory with Gyrodometry Method in the Indonesia 2022 Ash Robot Contest

Development of robot motion direction based on microcontroller with compass sensor

This research brings innovation to the motion and navigation system of the ‘DK-ONE’ robot. In the 2021 Indonesian ‘Search and Rescue’ robot contest, the ‘DK-ONE’ robot faced difficulties moving towards the target room. The issue was attributed to an unbalanced frame construction and friction between the robot’s legs and the arena floor, leading to leg slippage. This resulted in a mismatch between the programmed number of steps for the robot and the desired path to the target space, causing errors in the robot’s system. To address these problems, researchers conducted a study aimed at enabling the ‘DK-ONE’ robot to accurately determine its direction of motion. This research followed the waterfall method, involving stages such as system analysis, design, coding, testing, and supporting phases. The study was carried out in the integrated laboratory of the Department of Electrical Engineering Education. The development of the robot’s motion direction using a compass sensor significantly improved stability while walking on straight, flat, and uneven paths. The robot no longer experienced errors in its motion direction and remained on the intended path. As a result, the increased efficiency in robot motion also positively impacted the structural efficiency and energy consumption of the robot.

OPTIMASI SENSOR KOMPAS DENGAN METODE CIRCLE EQUATION PADA ROBOT SEPAK BOLA BERODA GERHANA DEWARUCI

The Indonesian Robot Contest (IRC) is an annual activity carried out by the National Achievement Center in the year of the change of government cabinet which was previously carried out by the Ministry of Research, Technology, and Higher Education. IRC has several competition categories, one of which is the Indonesian Wheeled Soccer Robot Contest. Most robots have accuracy problems in the navigation system during the game due to an error in the compass that is used to compare the gyroscope. As a result, many robots are inaccurate and even misdirected, both when attacking or defending. With these problems, one solution is to use the circle equation as a compass sensor filter, so that the robot can have an accurate navigation system, deviations from readings on the CMPS12 sensor can be reduced through the filtering process and have a large enough impact on the normalization process of CMPS12 sensor readings evenly. average error of 10° (degrees). With the addition of the circle equation method, the CMPS12 sensor reading data becomes more accurate with an average error of 1.08° (degrees). Compared with the kalman fillter method with CMPS12 sensor reading data which has an average error of 4 ° (degrees). In this research, the compass sensor is used to compare the gyroscope sensor by changing the gyroscope value using the reference reading of the CMPS12 value as a correction for gyroscope readings that are not in accordance with the set points and an average angle error of 0.6 ° (degrees) and 0.88 ° ( degrees) after the addition of the circle equation method. This method is better than the kalman fillter method with an average error of 3.1° (degrees) and 4.4° (degrees) but is better than that without the method with an average error of 3.6° (degrees) and 5 ,72° (degrees). The resulting error is smaller and can still be tolerated. CMPS12 as a gyroscope comparison can work well in this test because the wheeled KRSBI robot is designed so that the direction of the robot faces is appropriate, with a minimum error in each direction of motion.

Development of a Decommissioning Robot with a Simple Structure Capable of Traversing Steps Using Two Different Drive Systems

In response to the imperative need for robotic involvement in the decommissioning of the Fukushima Daiichi Nuclear Power Plant, a proactive initiative was undertaken. This initiative led to the organization of the “7th Decommissioning Creative Robot Contest,” in which pivotal challenges related to nuclear plant decommissioning were addressed. Recognizing inaccessible areas within a decommissioned plant, the primary objective of this study was to develop a remotely operated robot capable of executing decontamination tasks in such environments. Two distinct locomotion systems were designed: mecanum wheels and crawlers. The mecanum wheels, characterized by their ability to execute approximately 100 mm vertical movements, facilitated the robot to surmount obstacles, particularly by negotiating 100 mm steps by engaging the tracks. Simultaneously, to address decontamination tasks at elevated locations, a threefold expandable ladder truck system was integrated, augmenting the operational range of the robot. Participation in the contest provided a platform for showcasing the developed technologies and validating their practicality. Despite encountering challenges during competition, such as damage to one locomotion system, the other locomotion system can be used to reach the destination. Experience serves as a crucial means of verifying technological advancements. Issues such as malfunctions in the range of motion control due to faulty limit switch contacts and unexpected loads on the tire elevation motors were noted, which hindered the achievement of the anticipated results. Despite these drawbacks, the adopted structural design demonstrates notable advantages, particularly in terms of maneuverability. The combination of mecanum wheels and crawlers along with a ladder truck system demonstrated adaptability to various scenarios. This adaptability holds significant promise for future nuclear power plant decontamination applications. While the competition may not have yielded the desired outcome, the lessons learned and the verified practicality of the developed technologies reinforce their potential utility in addressing the future challenges of nuclear plant decommissioning.

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.

BCI Controlled Robot Contest on the 50th Anniversary of Brain-Computer Interfaces

BCI Controlled Robot Contest on the 50th Anniversary of Brain-Computer Interfaces

Penerapan Metode Fuzzy Logic-PID Sebagai Kontrol Gerak Robot Berkaki Enam (Hexapod)

Robotics technology is currently developing rapidly and has become one of the sophistications that are relied upon by every country, one of which is Indonesia. One of the activities for the development of robotics technology in Indonesia is the Indonesian Robot Contest (KRI). KRI was held as an event for students' creativity in designing robots. One of the divisions contested at KRI is the Indonesian Search and Rescue Robot Contest (KRSRI) division. In the KRSRI division, it is required to make a legged robot that can navigate the motion of walking in the competition area and pass obstacles on the track. Of course, a navigation motion control system is needed that can regulate the movement of the legged robot's steps properly, so that the robot can pass the path in the arena. In this study, the answer to this problem is the application of legged robot navigation motion control using the Fuzzy Logic-PID method. The Fuzzy Logic-PID method is a combination of two methods between the Fuzzy Logic Controller method and the Proportional Integral Derivative (PID) control method. The Fuzzy method’s role is to produce constant values ??P, I, and D based on the error value and the incoming error delta. Furthermore, the resulting constant value is used in the calculation of the PID control. From the experimental data, the Fuzzy method has been able to produce a PID constant value and the PID method can also be applied to regulate the robot's motion steps. The test results on trials using obstacles and without obstacles were successful, except for uphill obstacles with a success percentage of 60% successfully passing incline obstacles and 40% failing to pass uphill obstacles.

Implementation of Behavior Based Robotics on Hexapod Legged Robot Based on Room Mapping

The Indonesian Fire Fighting Robot Contest (KRPAI) has a mission to extinguish fires quickly, with various kinds of points and bonuses to be taken. The robot used in this research is a 6-legged robot (Hexapod). The problem faced today is the robot's ability to navigate the room, the robot still has to enter the same room, especially in room 1 (a room with 2 doors) so that the Arbitrary Start bonus or Variable Door location which prohibits the robot from entering the same room cannot be obtained. There is also a Room Factor bonus obtained using the JTS (Segment Distance) method which is an additional bonus when the robot walks outside the room. It is necessary to develop a room mapping algorithm with behavior based so that the robot can get maximum points. The main sensor used is the IMU BNO055 sensor as a robot direction orientation sensor, as well as a proximity sensor as a wall search sensor, so that the robot can know the direction when leaving and exploring the room, knowing the starting room, and ensuring the next room entered is a different room. The results of the test successfully made the robot take maximum points in the race.

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.

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.

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 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.

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.

Transformer-based ensemble deep learning model for EEG-based emotion recognition

Emotion recognition is one of the most important research directions in the field of brain–computer interface (BCI). However, to conduct electroencephalogram (EEG)-based emotion recognition, there exist difficulties regarding EEG signal processing; moreover, the performance of classification models in this regard is restricted. To counter these issues, the 2022 World Robot Contest successfully held an affective BCI competition, thus promoting the innovation of EEG-based emotion recognition. In this paper, we propose the Transformer-based ensemble (TBEM) deep learning model. TBEM comprises two models: a pure convolutional neural network (CNN) model and a cascaded CNN-Transformer hybrid model. The proposed model won the abovementioned affective BCI competition’s final championship in the 2022 World Robot Contest, demonstrating the effectiveness of the proposed TBEM deep learning model for EEG-based emotion recognition.

Junior High Robotics Inspiration: Engaging Students with Exciting Interactive Socialization

This study elucidates the impact of a dynamic robot-focused socialization endeavor on the students of SMP Muhammadiyah 4 Porong. Targeting grades 7, 8, and 9, the initiative aimed to bridge the educational void resulting from the school's relocation. The presentation encompassed diverse national and international junior-level robot contests, along with rudimentary insights into robot components and their functionalities, aligning with the students' foundational aptitudes. Practical involvement, including hands-on interaction with an android smartphone-controlled sumo robot, generated enthusiasm reminiscent of play. A noteworthy upshot was the emergence of approximately 10 female students displaying a newfound interest in robotics. This venture not only augmented general knowledge but also spurred curiosity and engagement, emphasizing the potential for broader academic implications in the realm of junior high school education.
 Highlights:
 
 Interactive Exposure: The socialization event introduced students to national and international robot contests, fostering awareness and igniting interest in the field.
 Hands-on Experience: Through direct interaction with a smartphone-controlled sumo robot, students were engaged in a playful yet educational manner, encouraging active participation.
 Gender-Inclusive Impact: The initiative notably attracted around 10 female students, potentially contributing to gender diversity and inclusivity in the realm of robotics education.
 
 Keywords: Robotics, Junior High School, Socialization, Engagement, Sumo Robot

Implementation of Grid Mapping Method for Firefighting Legged Robot

The firefighting robot contest is an event to encourage high education students to explore electronics and robotic skill. The difficulty level of the contest increases from year to year though the arena layout is relatively similar. The field of arena consists of maze, detachable doors and distractorswhich requires smart idea to enable the robot to be well-performed. The combination of home position and random fire spot at each round of the game make each team have to find the most effective method to win the match. This paper presents an implementation the grid map method on the Azmi four legged robot using ultrasonic PING sensor, GPYA021 infrared sensor and GY-955 gyroscope. An ultrasonic sensor is attached on each of four sides of the robot. The data from each sensor will be interpreted as the distance of the front, right, left and back of the robot relative to the wall. Interpretation depends on the robot's current position. Virtual mapping technique was used by assuming that the arena consists of grids which have uniform size. Each grid was given a consecutive index numbering i.e a symbolic number which in practice be manifested in terms of distance unit. The results showed that the grid map method in this study worked well, tested on four randomized configurations, the robot successfully carried out the search task and returned to the home position with an acceptable execution time.

Development of a Robot for Decontamination of High Places and Decommissioning Work That Can Cope with Slopes and Steps

In this study, a robot was created to compete in the “7th Decommissioning Creative Robot Contest. The “Decommissioning Creative Robot Contest” is a robot contest in which the fields and tasks are set based on the assumption of decommissioning work at TEPCO’s Fukushima Daiichi Nuclear Power Station. The theme of this year’s competition is 3D decontamination of a highly contaminated area inside a nuclear reactor building. To solve the competition problem, we propose a robot for decommissioning work equipped with “an undercarriage mechanism that can cope with slopes and steps”, “a decontamination mechanism with two types of rotation trajectories”, and “a lifting mechanism with a high deformation rate. The undercarriage mechanism uses a four-wheel drive with infinite orbit to cope with slopes and steps. The decontamination mechanism uses two different types of orbits for decontamination: a circular rotating orbit and a longitudinal rotating orbit for rotation. For the lifting mechanism, three pantograph mechanisms were used. This study realized a control and mechanism to overcome steps and slopes, a deployment mechanism to high places, and a mechanism with two types of orbits to carefully decontaminate a large area.

Implementasi Fuzzy Logic untuk Kontrol Gerak Tari pada Robot Humanoid dengan Menerapkan Sensor Deteksi Warna TCS3200

One of the prominent competitions in the national robot series, the Indonesian dancing Robot Contest (KRSTI), assesses the dependability of various dancing robots by having them perform traditional dances from Indonesia. In this study, the authors use fuzzy logic to control the mobility of the dancing robot by determining whether it poses and dances in accordance with the KRI dance zone based on the input produced by the TCS3200 color sensor. In order to create a robot that may move in accordance with its designated zone, the Mamdani fuzzy approach, which employs the IF-ELSE principle, is used for decision-making.

The Natural Robotics Contest: crowdsourced biomimetic design

Biomimetic and bioinspired design is not only a potent resource for roboticists looking to develop robust engineering systems or understand the natural world. It is also a uniquely accessible entry point into science and technology. Every person on Earth constantly interacts with nature, and most people have an intuitive sense of animal and plant behaviour, even without realizing it. The Natural Robotics Contest is novel piece of science communication that takes advantage of this intuition, and creates an opportunity for anyone with an interest in nature or robotics to submit their idea and have it turned into a real engineering system. In this paper we will discuss the competition’s submissions, which show how the public thinks of nature as well as the problems people see as most pressing for engineers to solve. We will then show our design process from the winning submitted concept sketch through to functioning robot, to offer a case study in biomimetic robot design. The winning design is a robotic fish which uses gill structures to filter out microplastics. This was fabricated into an open source robot with a novel 3D printed gill design. By presenting the competition and the winning entry we hope to foster further interest in nature-inspired design, and increase the interplay between nature and engineering in the minds of readers.