An Intelligent Ground-Air Cooperative Navigation Framework Based on Visual-Aided Method in Indoor Environments

Abstract

In the future, heterogeneous robots are expected to perform more complex tasks in a cooperative manner, and the onboard navigation system is required to be capable of working safely and effectively in the area where GNSS signal is weak or even could not be received. To demonstrate this concept, we have developed a cooperative navigation system by the use of Ground-Aerial Vehicle Cooperation. The key innovations of the development lie in the following aspects: (1) a local scalable self-organizing network is constructed for data interaction between a small UAV and a reusable ground robot; (2) a new navigation framework is proposed to achieve visual simultaneous localization and mapping (SLAM) where carrying capacity of both the ground vehicle and UAV are systematically considered; (3) an octomap-based 3D environment reconstruction method is developed to achieve map pre-establishment in complex navigation environments, and the classic ORB-SLAM2 system is improved to be adaptive to actual environment exploration and perception. In-door flight experiments demonstrate the effectiveness of the proposed solution. More interestingly, by implementing a centroid tracking algorithm, the cooperative system is further capable of tracking a man moving in indoor environments.