Visual Navigation Technology for Autonomous Vehicle

我们已经开发了一种新的多视觉传感器（MISS），它结合了全向图像传感器COPIS（CONIC投影图像传感器）和双目视觉，用于导航机器人和理解感兴趣的物体在环境中，通过集成两个传感器数据。由于COPIS观察机器人周围的360度视图，因此在机器人的已知运动的假设下，通过监视图像中的方位角变化来生成场景的环境地图。我们在室外环境中做了几个实验。所获得的环境地图的精度是足够的机器人导航在这样的环境。在这项工作中，我们通常提取垂直边缘的假设下，完美的地平面运动的机器人。然而，如果机器人由于地平面的不均匀性而摇动，则有必要估计滚动和摇摆运动并稳定图像。全向图像传感器中的光流径向分量具有周期性，且不随机器人地平面运动而变化。因此，利用这些特性，我们的方法可以估计滚动和摇摆运动通过拟合的径向和分量的光流周期函数模型，并通过评估光流的圆周分量的对称性。另一方面，移动的机器人具有双目视觉，并获得一系列立体图像的环境。虽然双目视觉的视野受到透镜的视角的限制，但是双目视觉对于理解环境的空间配置是有用的。因此，我们综合两者的优点，并提出了一个有效的传感系统。在实际环境中对原型传感器进行了评估，并提出了一种在动态变化的三维环境中障碍物运动无法预测的实时路径规划算法。

We have developed a new multiple visual sensing sensor (MISS), which combines with an omnidirectional image sensor COPIS (COnic Projection Image Sensor) and binocular vision, for navigating the robot and understanding interesting objects in an environment by integrating both sensory data. Since COPIS observes a 360 degree view around the robot, under the assumption of the known motion of the robot, an environmental map of an scene is generated by monitoring azimuth change in the image. We did several experiments in the outdoor environment. The precision of obtained environmental maps was sufficient for robot navigation in such environment. In this work, we usually have extracted vertical edges under the assumption of perfect ground plane motion of the robot. However, if the robot joggles due to unevenness in the ground plane, it is necessary to estimate the rolling and swaying motion and stabilize the image. The radial component of optical flow in omnidirectional image sensor has a periodic characteristic and is invariant to the ground plane motion of the robot. Therfore, By making use of these characteristics, our method can estimate the rolling and the swaying motions by fitting the radial and component of optical flow to a periodic functional model and by evaluating symmetry of circumferential component of optical flow.On the other hand, The mobile robot has binocular vision and obtains a sequence of stereo images from the environment. Although a view field of the binocular vision is limited by a visual angle of lens, the binocular vision is useful for understand spatial configuration of the environment. Therefore we integrate both merits and propose an efficient sensing system. The system has been evaluated on the prototype sensor in actual environment.Furthermore, we have proposed a real-time algorithm for path planning in a dynamically changing 3-D environment where the motion of the obstacles can not be predicted.