This is a paper recently accepted by IEEE-ASME Transactions on Mechatroncis, finished by my post graduate student Yajun Gao.
Abstract: Stable communication plays an important role in many autonomous robotic applications, particularly in challenging environments that lack communication infrastructures. At present, there exists a high demand for wireless communication networks that can be quickly established to operate devices or agents, such as tasked mobile robots. Eliminating the bottlenecks in data transmission and optimizing the communication ability of base stations, mobile robots, and clients are necessary to ensure stable wireless communication. To solve the problem, an efficient approach is proposed by equipping Wi-Fi routers on mobile robots to enable and enhance dynamic communication ability. First, two specific Wi-Fi models are proposed to detect the distribution of Wi-Fi signals over operating environments and assist in the navigation of relay robots. A visual-laser simultaneous localization and mapping is proposed to establish an environmental map and further localize relay robots. A rapidly-exploring random trees-based motion planning method is utilized to identify the target relay locations with optimal communication ability based on the built Wi-Fi signal distribution. Mobile relay robots are controlled automatically to the corresponding target locations, and an ad-hoc wireless network with good quality is established. The experimental results are presented to demonstrate the effectiveness of the proposed approach.
The experimental video can be found here.