基于IA-DWA算法的移动机器人的自主导航研究
Quanling He1,2, Zongyan Wang3,4, Kun Li1
1North University of China, School of Mechanical Engineering, Taiyuan, 030051, Shanxi, China.
Scientific reports
|January 15, 2025
概括
本研究介绍了移动机器人改进的路径规划算法 (IA-DWA),提高了效率和避开障碍. 与现有方法相比,新方法可减少23.3%的规划时间和1.8%的路径长度.
科学领域:
- 机器人技术 机器人技术 机器人技术
- 人工智能的人工智能
- 导航系统 导航系统
背景情况:
- 移动机器人导航需要高效准确的路径规划,以避免障碍.
- 像A*和动态窗口方法 (DWA) 这样的现有算法在速度和全球最佳性方面存在局限性.
- 准确的机器人定位对于可靠的路径执行至关重要.
研究的目的:
- 开发一个综合算法 (IA-DWA),将A*和DWA结合起来,以改进移动机器人路径规划.
- 为了提高速度,效率和无碰撞路径生成能力.
- 为了提高机器人的定位准确度,使用传感器融合.
主要方法:
- 用扩展的卡尔曼波器 (EKF) 进行精确的机器人定位的融合式公里表和惯性测量单元 (IMU) 数据.
- 优化了A*算法参数 (预测函数,权重,邻近,平滑) 以将全球路径信息集成到DWA中.
- 在机器人操作系统 (ROS) 移动机器人平台上模拟和实验验证IA-DWA算法.
主要成果:
- 与A*-DWA.相比,IA-DWA算法减少了23.3%的路径规划时间.
- 使用IA-DWA,路径长度减少了1.8%,表明效率有所提高.
- 观察到优化代的更快的趋同.
- 实验验证证了算法在自主导航中的可靠性.
结论:
- 拟议的IA-DWA算法显著提高了移动机器人路径规划的效率和准确性.
- 传感器融合 (EKF) 增强了机器人的定位,这对于现实世界导航至关重要.
- 综合方法有效地平衡了全球路径的最佳性与避免局部障碍.
相关概念视频
01:29Relative Motion Analysis using Rotating Axes-Problem Solving
Consider a crane whose telescopic boom rotates with an angular velocity of 0.04 rad/s and angular acceleration of 0.02 rad/s2. Along with the rotation, the boom also extends linearly with a uniform speed of 5 m/s. The extension of the boom is measured at point D, which is measured with respect to the fixed point C on the other end of the boom. For the given instant, the distance between points C and D is 60 meters.
Here, in order to determine the magnitude of velocity and acceleration for point...
Here, in order to determine the magnitude of velocity and acceleration for point...
01:30Vector Functions and Motion: Problem Solving
Accurate position tracking is fundamental to the safe and effective operation of unmanned aerial vehicles (UAVs), particularly during precision maneuvers near complex structures. In this scenario, a drone is programmed to perform a high-precision inspection of a vertical structure, starting at position ((x, y, z) = (3, 0, 0)), with an initial velocity oriented in the positive z-direction. The trajectory of the drone is governed by a time-dependent acceleration function a(t), which is predefined...