本文为美国加州理工学院（作者：Jeremy Ma）的博士论文，共120页。

机器人感知是任何自主系统的一个基本方面，它使机器人能够理解大量的数据，并了解周围的世界。三维物体的实时检测和位姿估计是机器人感知领域的一个热点问题。本文提出了一种利用立体相机传感器进行三维目标检测与跟踪的方法。通过训练阶段在短时间内学习几何对象模型，并通过在相机图像的自适应感兴趣区域内对所选特征执行稀疏立体计算，使实时检测和跟踪成为可能。实验结果表明，该方法与实时的真实测量方法相比，是一种有效的方法。以该框架为基础，考虑了机器人传感中两个其他问题的扩展：（1）用于模型识别的传感器规划；（2）用于目标搜索的传感器规划。在前者中，提出了一种新的确定移动传感器从已知模型数据库中识别未知三维物体的新算法，并通过两个实际机器人系统的实验进行了测试。采用信息论的方法量化了每种潜在的感知行为的效用，并对算法的有效性进行了讨论。在后一个领域，提出了一种新的方法，允许自主移动机器人使用安装在云台上的车载立体摄像机传感器来搜索三维物体。搜索效率由粗尺度全局搜索与精尺度局部搜索决定，由基于网格的概率图引导。将搜索过程中的避障问题自然地融入到该方法中，并在移动机器人上进行了实验，以说明和验证所提出的搜索策略。所有的实验都是在一台笔记本电脑上进行的，并且计算量不大。

Robot perception is a fundamental aspect of any autonomous system. It gives the robot the capacityto make sense of vast amounts of data and gain an understanding of the world around it. An activeproblem in the area of robot perception is real-time detection and pose estimation of 3D objects.This thesis presents an approach to 3D object detection and tracking utilizing a stereo-camerasensor. Geometric object models are learned in short order time via a training phase and real-time detection and tracking is made possible by performing sparse stereo calculations on the chosenfeatures within an adaptive region of interest of the camera image. The experimental results obtainedby using this method will show the effectiveness of the approach as compared against ground truthmeasures in real-time. Using that framework as a basis, extensions to two other problems in robotsensing are then considered: (1) sensor-planning for model identification, and (2) sensor-planningfor object-search. In the former, a novel algorithm for determining the next-best-view for a mobilesensor to identify an unknown 3D object from among a database of known models is presented andtested across two experiments involving real robotic systems. An information theoretic approachis taken to quantify the utility of each potential sensing action and the validity of the algorithm isdiscussed. In the latter area, a novel approach is presented that allows an autonomous mobile robotto search for a 3D object using an onboard stereo camera sensor mounted on a pan-tilt head. Searchefficiency is realized by the combination of a coarse-scale global search coupled with a ?ne-scale localsearch, guided by a grid-based probability map. Obstacle avoidance during the search is naturallyintegrated into the method with additional experimental results on a mobile robot presented toillustrate and validate the proposed search strategy. All presented experiments were carried out inreal-time processing with modest computation done by a single laptop computer.

1.  引言

2.  项目背景

3.  3D位姿估计的目标检测

4.  模型辨识的传感器规划

5.  SPPEMI实现

6.  3D目标搜索的传感器规划

7.  结论与展望

附录A 线性测量矩阵

附录B 无约束运动的运动模型

更多精彩文章请关注公众号：