This article focuses on cyber-physical systems involving a set of robots and sensors that are interconnected through wireless links. In applications that operate in harsh environments, wireless sensors and robot networks (WSRNs) may suffer simultaneous failures, which lead to partitioning the WSRN into unconnected subnetworks. The loss of connectivity often leads to degrade performance and inability to fulfill the application mission. To address this issue, several proposals have been made in the literature where node mobility is exploited to re-establish connectivity. These proposals, however, focus on minimizing the number of mobile nodes and the traveled distance cost while ignoring important factors, such as robot battery level, network lifetime, and environmental conditions, that may limit or hinder the whole recovery process. To overcome these limitations, we present a distributed node repositioning algorithm based on fuzzy logic. We evaluate our approach via extensive simulations and prototype experiments using Khepera IV robots. The results demonstrate superior performance to that of contemporary approaches.