The great development of the robotics design and production make the robots be an important part in the achievement of many of the difficult or dangerous applications to humans, one of these robots are the climbing robots. The design of climbing robots is a major challenge, because the robot has to stick to the walls while the walls are different in terms of roughness. In this paper, the arm of climbing robot has been designed to help robot climbs on coarse surfaces where a gecko arm model is used to achieve linear movement while clinging to the rough wall achieved by using limb has claws as the limbs of cats. A mathematical model has been derived and simulated by MATLAB while the mechanical parts were constructed using plastic materials and motion has been achieved by servo motors, also microcontroller kit used to control the arm and achieving motion synchronization. Several experiments have been performed in order to test the success of the arm of climbing robot.
As robots are expected to get more involved in people’s everyday lives, frameworks that enable intuitive user interfaces are in demand. Hand gesture recognition systems provide a natural way of communication and, thus, are an integral part of seamless human-robot interaction (HRI). Recent years have witnessed an immense evolution of computational models powered by deep learning. However, state-of-the-art models fall short of expanding across different gesture domains, such as emblems and co-speech. In this paper, we propose a novel hybrid hand gesture recognition system. Our Snapture architecture enables learning both static and dynamic gestures: by capturing a so-called snapshot of the gesture performance at its peak, we integrate the hand pose and the dynamic movement. Moreover, we present a method for analyzing the motion profile of a gesture to uncover its dynamic characteristics, which allows regulating a static channel based on the amount of motion. Our evaluation demonstrates the superiority of our approach on two gesture benchmarks compared to a state-of-the-art CNNLSTM baseline. Our analysis on a gesture class basis unveils the potential of our Snapture architecture for performance improvements using RGB data. Thanks to its modular implementation, our framework allows the integration of other multimodal data, like facial expressions and head tracking, which are essential cues in HRI scenarios, into one architecture. Thus, our work contributes both to integrative gesture recognition research and machine learning applications for non-verbal communication with robots.
Current consumer virtual reality (VR) systems rely heavily on handheld controllers for input. To this end, numerous methods have been developed and investigated since the emergence of VR in order to improve the user experience for interactions such as gaming or text entry while wearing a head-mounted display (HMD) and using handheld controllers [1, 2]. Although several novel text entry methods have been proposed, there is little research comparing the methods using virtual keyboard interfaces evaluated as experimental conditions. This would allow for a more focused and specialized comparison.Statement of Objective:This work presents the design and empirical evaluation of a split and standard virtual keyboard for text input in virtual environments using handheld controllers. There are numerous applications, including games, for example, entering a gamer’s name or messaging other gamers. An experimental evaluation of two virtual keyboards using VR handheld controllers will be conducted. (Note: As the experiment is ongoing, this abstract is written in the future tense. The final submission will transition to the past tense.) The focus will be on entry speed, accuracy, and efficiency. Both keyboards will be QWERTY-based, but with different organizations. Similar to most physical keyboards, the “standard” keyboard will have all the keys in one arrangement. However, the “split” keyboard will display keys in a split pattern on each side of the screen. For the standard keyboard, users will operate one handheld controller in their preferred hand. For the split keyboard, users will operate one controller in each hand. Thus, a significant point of comparison in the present research is one-handed input vs. two-handed input for the same task. Description of Methods:This research follows an experimental methodology. There will be a total of 14 participants recruited from the local university campus. The study will be a 2 × 5 within-subjects design with the following independent variables and levels:• Keyboard (standard one-handed, split two-handed)• Block (1, 2, 3, 4, 5)The order of testing the keyboards will be counterbalanced to offset learning effects.Five phrases of text will be entered for each block, with phrases selected at random from a standard phrase set [3]. Thus the total number of trials will be 14 participants x 2 keyboards x 5 blocks x 5 phrases/block = 700. The following are the dependent variables in the study:• Entry speed (wpm)• Error rate (%)• Keystroke per character (KSPC)The Unity 3D game engine is used to create the text-input interface installed on the Meta Quest 2 VR headset. Participants will use the Meta Quest handheld controllers to input text. Input uses raycasting, in which the handheld controller casts a virtual line or ray. The virtual ray is positioned on a key on the virtual keyboard, wherein the user presses the trigger on the handheld controller to select the character indicated by the ray.Significance of the Proposed Presentation:The work is significant due to its presentation of the design and experimental evaluation of novel text entry input methods for virtual environments The work pushes the limits of contemporary devices (Meta Quest 2 VR headset) and platforms (Unity 3D game engine) into a design space of particular interest to research in virtual environments.Discussion of Results:Statistical tests, such as the analysis of variance, will be used to identify statistically significant differences between the keyboards for entry speed (wpm), error rate (%), and efficiency using KSPC. Tests for improvement in performance over the five blocks of testing will be included, as well. Regression models will be built using the power law of practice to model the pattern of learning over the blocks of testing. Design implications and suggestions will be offered, as will opportunities for future work. Full details are to be included in the final submission.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.