Coordinated control and navigation of field robots

本提案申请资金用于购买研究移动机器人平台，以促进现场机器人的研究。研究目标是在纪念大学开发一种现场机器人能力，以推进和实施自主机器人技术，用于需要机器人在更大的区域（或距离）中导航的应用，可以适用于室内和室外条件。在现场机器人技术下，预计将在几个领域开展研究，即基于任务特定行为的机器人控制和协调，同时定位和绘图（SLAM）技术，以覆盖更大的距离（约数百公里），生物启发的控制方法，用于使用领导者-追随者技术的编队控制，最后是多机器人监视与最优路径规划。机器人的行为调节包括在单一目标下操作多个机器人的行为形成，并将其转化为一系列导航任务，以产生在非结构化环境中保持队形的运动控制命令。适用于机器人的约束包括噪声和有限的传感器感知，机器人状态传输的通信能力有限，最后是环境的约束。第二个问题是在更大范围的测绘和定位背景下SLAM技术的发展。第三个也是最后一个问题是发展机器人在人类监督下操作的认知能力。在这种情况下，基于多模态传感器的机器人注意力技术将在人类监督下得到发展。

This proposal requests funding to purchase a research mobile robot platform to facilitate research in field robotics. The research objective is to develop a field robotic capability in Memorial University to advance and implement autonomous robotic technology for applications that requires robots to navigate through a larger area (or a distance) that can be applicable for both indoor and outdoor conditions. Under field robotics it is expected to develop research in several areas, namely, task specific behaviour based control and coordination of robots, simultaneous localization and mapping (SLAM) techniques to cover a larger distance (in the order of hundred kilometers), biologically inspired control methods for formation control using leader-follower techniques and finally multi-robot surveillance with optimal path planning. The behaviour modulation for robots includes formation of behaviours to operate multiple robots under a single objective and translation them in to a sequence of navigational tasks to produce motor control commands for keeping a formation in an unstructured environment. The constraints applicable for the robots include noisy and limited sensor perception, limited communication capabilities in transmitting robot state and finally constraints imposed by the environment. The second issue is related to the development of SLAM techniques in the context of larger area of mapping and localization. The third and final issue is related to develop cognitive capability for robots to operate with human supervision. In this case the multi-modal sensor based robotic attention techniques will be developed to work with human supervision.