ABSTRACT

Navigation is an essential ability for mobile agents to be completely autonomous and able to perform complex actions. However, the problem of navigation for agents with limited (or no) perception of the world, or devoid of a fully defined motion model, has received little attention from research in AI and Robotics. One way to tackle this problem is to use guided navigation, in which other autonomous agents, endowed with perception, can combine their distinct viewpoints to infer the localization and the appropriate commands to guide a sensory deprived agent through a particular path. Due to the limited knowledge about the physical and perceptual characteristics of the guided agent, this task should be conducted on a level of abstraction allowing the use of a generic motion model, and high-level commands, that can be applied by any type of autonomous agents, including humans. The main task considered in this work is, given a group of autonomous agents perceiving their common environment with their independent, egocentric and local vision sensors, the development and evaluation of algorithms capable of producing a set of high-level commands (involving qualitative directions: e.g. move left, go straight ahead) capable of guiding a sensory deprived robot to a goal location. In order to accomplish this, the present paper assumes relations from the qualitative spatial reasoning formalism called StarVars, whose inference method is also used to build a model of the domain. This paper presents two qualitative-probabilistic algorithms for guided navigation using a particle filter and qualitative spatial relations. In the first algorithm, the particle filter is run upon a qualitative representation of the domain, whereas the second algorithm transforms the numerical output of a standard particle filter into qualitative relations to guide a sensory deprived robot. The proposed methods were evaluated with experiments carried out on a 2D humanoid robot simulator. A proof of concept executing the algorithms on a group of real humanoid robots is also presented. The results obtained demonstrate the success of the guided navigation models proposed in this work.

摘要

导航是移动代理完全自治并能够执行复杂动作的基本能力。但是，对于具有有限（或没有）世界感知或没有完整定义的运动模型的特工的导航问题，在AI和机器人学方面的研究很少受到关注。解决此问题的一种方法是使用导航，其中具有感知能力的其他自治主体可以结合其独特的观点来推断定位，并通过适当的命令来引导感官匮乏的主体通过特定路径。由于对被引导代理的物理和感知特性的了解有限，因此应在抽象级别上执行此任务，以允许使用通用运动模型和高级命令，可以由任何类型的自治代理（包括人类）应用。在这项工作中考虑的主要任务是，考虑到一组自治代理通过其独立的，以自我为中心的和本地的视觉传感器感知其共同的环境，开发并评估能够生成一系列高级命令（涉及定性方向）的算法：例如向左移动，一直往前走）能够引导被剥夺的感官机器人到达目标位置。为此，本文从定性的空间推理形式主义（称为StarVars）中假设关系，该推理方法还用于构建领域模型。本文提出了两种使用粒子滤波器和定性空间关系的制导导航的定性概率算法。在第一种算法中，粒子过滤器在域的定性表示上运行，而第二种算法将标准粒子过滤器的数值输出转换为定性关系，以指导感官匮乏的机器人。通过在2D人形机器人模拟器上进行的实验对提出的方法进行了评估。还提供了在一组真实的类人机器人上执行算法的概念证明。