Scholarlink

Traditional robotic control systems, relying on complex joysticks and control panels, are often a significant barrier to user adoption. In this paper, we introduce a natural, gesture-based interface for robotic vehicle control through intuitive hand gestures. Our proof-of-concept uses computer vision, a custom convolutional neural network (CNN), and wireless communication to identify these gestures and translate them into real-time robot action. The system provides a controlled gesture recognition rate of 90% and demonstrates strong performance in a range of real-world scenarios. One of the distinguishing features of our system is the multi-purpose robotic arm, enabling the vehicle to perform manipulation tasks and navigation. We established the practicality of the system with extensive indoor and outdoor trials, confirming its applicability to use cases that span from assistive technology to industrial automation. Our work envisions a future of more cooperative and accessible human-robot interactions by swapping a widely used form of communication for complex controls.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Intelligent Hand Gesture Recognition Using Convolutional Neural Networks for Wireless Robotic Vehicle Control

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Rakhi Bharadwaj,
  • Kushal Chaudhari,
  • Satyajit Bhandari,
  • Ishika Mahajan,
  • Ruhi Tendulkar,
  • Vedant Gulhane,
  • Sujit Kanawade

摘要

Traditional robotic control systems, relying on complex joysticks and control panels, are often a significant barrier to user adoption. In this paper, we introduce a natural, gesture-based interface for robotic vehicle control through intuitive hand gestures. Our proof-of-concept uses computer vision, a custom convolutional neural network (CNN), and wireless communication to identify these gestures and translate them into real-time robot action. The system provides a controlled gesture recognition rate of 90% and demonstrates strong performance in a range of real-world scenarios. One of the distinguishing features of our system is the multi-purpose robotic arm, enabling the vehicle to perform manipulation tasks and navigation. We established the practicality of the system with extensive indoor and outdoor trials, confirming its applicability to use cases that span from assistive technology to industrial automation. Our work envisions a future of more cooperative and accessible human-robot interactions by swapping a widely used form of communication for complex controls.