We propose to enable high-throughput, scalable communication for cooperative autonomous vehicles through zero mutual interference (ZMI) networks. Cooperation in complex environments relies on wireless communication, but conventional wireless networks are not designed for cooperative autonomous vehicles and are fundamentally limited by mutual interference. Our approach is to avoid mutual interference by design; ZMI networks provide wired network properties using wireless radio links. In this paper, we present the initial instantiation of a ZMI network based on a multi-radio, multichannel architecture. The network is constructed such that each vehicle communicates with topological neighbours using a dedicated radio and channel. We also present experimental results that compare the performance of a ZMI network to a conventional inter-vehicle communication network in a cooperative perception and control task.

Decentralized control of modular robots requires reliable inter-module communication. Communication links must tolerate module misalignment and implement the neighbor-to-neighbor communication model. We propose a wireless system based on multiple radios per module that addresses these challenges. Our system scales to large robots because available bandwidth is independent of the number of modules. In this paper, we present our multi-radio single-channel architecture and validate its performance through hardware experiments. Results show that radios can provide reliable neighbor-to-neighbor communication suitable for modular robots.