基于粒子群优化的机器人多传感器自标定方法

A Robot Multi-sensor Self-calibration Method Based on Particle Swarm Optimization

  • 摘要: 针对机器人感知系统中多个传感器的位姿关系估计问题,提出一种新的基于粒子群优化的自标定方法.利用激光扫描点在图像上的投影线约束条件,建立摄像机与激光扫描仪外参数的非线性优化目标函数,采用粒子群算法优化外参数.该方法对参数的初值并不敏感,无需特定的标定物.实验结果表明,该算法的标定精度高.

     

    Abstract: Aiming at the estimation problem of the position and orientation relationship of multiple sensors in robotic perceptual system,a new self-calibration approach based on particle swarm optimization(PSO)is proposed.According to the constraint of the projection line produced by laser scanning points in image,the nonlinear optimization objective function of the extrinsic parameters between a camera and a laser scanner is established.And particle swarm optimization method is adopted for extrinsic parameters.The approach is insensitive to the initial value of parameters,and does not require any calibration object as well.The experimental results confirm that the proposed algorithm can yield high accurate calibration results.

     

/

返回文章
返回