New approaches to intelligent navigation of autonomous surface and underwater robotic systems

无人水面机器人(USR)和无人水下机器人(UUR)在河流、湖泊和海洋中的安全监控、环境监测和勘探等方面有着重要的应用，特别是在非结构化和危险环境中。实时感知、定位和导航是自主水面机器人(ASR)和自主水下机器人(AUR)的基本问题。关键的挑战是向这些机器人注入“类似人类”的智能，使其能够安全自主导航，并成功完成指定任务。 拟议的5年研究计划是利用生物启发的智能方法，研究自主水面和水下机器人系统的实时传感、定位、路径规划、跟踪和控制。重点研究内容包括：(1)水面机器人实时无碰撞导航与控制，包括实时传感与多传感器融合、船上传感器信息与其他传感信息的融合、同步定位与测绘(SLAM)、水下机器人实时无碰撞路径生成与控制；(2)存在未知洋流和其他不确定性的动态三维水下环境中自主水下机器人的实时无碰撞导航与控制。将开发水面和水下机器人实验平台作为本研究的试验台。受生物启发的新型智能方法将用于水面和水下机器人系统的多传感器融合和自主导航，并将在实验平台上实现。将参照已有的领先方法，对新方法进行严格的理论研究和实验验证。 提出这项研究的动机是，鱼类和其他动物等生物系统可以有效地感知动态环境，并根据这些输入采取有效行动。拟议研究的新颖性和预期意义将是向各自的研究界提供新的知识，并开发受生物启发的新的智能技术，用于自主水面和水下机器人的实时传感和导航，这将有许多民用和军事应用。此外，拟议的研究将为其他研究领域提供一些关于生物启发方法的见解。此外，这一研究计划将为培训机器人和智能系统领域的高素质人才提供良好的机会，并产生巨大的商业和社会影响。

Unmanned surface robots (USR) and unmanned underwater robots (UUR) have many important applications such as security surveillance, environmental monitoring and exploration in rivers, lakes and seas, particularly in unstructured and hazardous environments. Real-time sensing, localization, and navigation are fundamentally important issues for autonomous surface robots (ASR) and autonomous underwater robots (AUR). The key challenge is to infuse these robots with “human-like” intelligence that will enable safe autonomous navigation and successful completion of designated tasks. The proposed 5-year research program is to investigate the real-time sensing, localization, path planning, tracking, and control of autonomous surface and underwater robotic systems, using biologically inspired intelligent approaches. In particular, the proposed research will focus on: (1) real-time collision-free navigation and control of water surface robots, including real-time sensing and multi-sensor fusion, integration of acquired information from its onboard sensors and other available sensed information, simultaneous localization and mapping (SLAM), and real-time collision-free path generation and control of the water surface robots; and (2) real-time collision-free navigation and control of autonomous underwater robots in dynamic 3-dimensional (3D) underwater environments, with the presence of unknown ocean currents and other uncertainties. Experimental surface and underwater robotic platforms will be developed as a test bed for this research. Novel biologically inspired intelligent approaches to multi-sensor fusion and autonomous navigation of surface and underwater robotic systems will be developed and implemented on the experimental platforms. Rigorous theoretical investigation and experimental verification of the novel approaches will be conducted, with reference to established leading approaches. The proposed research has been motivated by the fact that biological systems, such as fishes and other animals, can efficiently sense the dynamic environments and take effective actions based on those inputs. The novelty and anticipated significance of the proposed research will be the provision of new knowledge to the respective research communities and the development of new biologically inspired intelligent techniques for real-time sensing and navigation of autonomous surface and underwater robots, which will have many civilian and military applications. In addition, the proposed research will offer some insights into biologically inspired methodologies for other areas of research. Furthermore, this research program will provide good opportunities for the training of highly qualified personnel in the areas of robotics and intelligent systems, and also offer great commercial and social impacts.