Scholarlink

<p>This study presents the development and comprehensive evaluation of an autonomous ground vehicle prototype designed specifically for the diverse and complex road conditions prevalent in India. The prototype incorporates a multi-sensor suite, including RP LIDAR A1M8, Intel RealSense Depth Camera, ultrasonic distance sensor HC-SR04, GPS sensor neo6m, and 1D LIDAR TF LUNA micro, integrated with advanced control algorithms. Steering control utilizes a path-tracking algorithm based on a kinematic bicycle model, while acceleration is managed by a PID controller. The system’s performance was rigorously tested across various scenarios, encompassing different weather conditions, road types, curvatures, and obstacle densities. Results indicate a maximum accuracy of 98.2% under optimal conditions, decreasing to 80% in rainy weather and 79.1% on roads with potholes and uneven surfaces. The prototype demonstrated varying levels of accuracy with increasing road curvature and obstacle density. This research provides valuable insights into the challenges and potential solutions for autonomous vehicle operation in the unique context of Indian road environments, highlighting areas for future improvement in sensor technology and control algorithms to enhance safety and reliability in diverse conditions.</p>

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

Evaluation of Path-Planning Algorithms for Autonomous Navigation in Unstructured Indian Road Environments Considering the Effects of Weather and Road Curvatures

  • 期刊论文

  • 发表日期：

    2025年6月1日

查看全文
  • Vivek Patkar,
  • Prachi Bhoir,
  • Ninad Dileep Mehendale

摘要

This study presents the development and comprehensive evaluation of an autonomous ground vehicle prototype designed specifically for the diverse and complex road conditions prevalent in India. The prototype incorporates a multi-sensor suite, including RP LIDAR A1M8, Intel RealSense Depth Camera, ultrasonic distance sensor HC-SR04, GPS sensor neo6m, and 1D LIDAR TF LUNA micro, integrated with advanced control algorithms. Steering control utilizes a path-tracking algorithm based on a kinematic bicycle model, while acceleration is managed by a PID controller. The system’s performance was rigorously tested across various scenarios, encompassing different weather conditions, road types, curvatures, and obstacle densities. Results indicate a maximum accuracy of 98.2% under optimal conditions, decreasing to 80% in rainy weather and 79.1% on roads with potholes and uneven surfaces. The prototype demonstrated varying levels of accuracy with increasing road curvature and obstacle density. This research provides valuable insights into the challenges and potential solutions for autonomous vehicle operation in the unique context of Indian road environments, highlighting areas for future improvement in sensor technology and control algorithms to enhance safety and reliability in diverse conditions.