The Internet of Things (IoT) is on the verge of a major paradigm shift. In the IoT system of the future, IoFT, the “cloud” will be substituted by the “crowd” where model training is brought to the edge, allowing IoT devices to collaboratively extract knowledge and build smart analytics/models while keeping their personal data stored locally. This paradigm shift was set into motion by the tremendous increase in computational power on IoT devices and the recent advances in decentralized and privacy-preserving model training, coined as federated learning (FL). This article provides a vision for IoFT and a systematic overview of current efforts towards realizing this vision. Specifically, we first introduce the defining characteristics of IoFT and discuss FL data-driven approaches, opportunities, and challenges that allow decentralized inference within three dimensions: (i) a global model that maximizes utility across all IoT devices, (ii) a personalized model that borrows strengths across all devices yet retains its own model, (iii) a meta-learning model that quickly adapts to new devices or learning tasks. We end by describing the vision and challenges of IoFT in reshaping different industries through the lens of domain experts. Those industries include manufacturing, transportation, energy, healthcare, quality & reliability, business, and computing.
Steer-by-wire (SBW) systems provide significant benefits to classical vehicles, and are indispensable for fully autonomous vehicles (AVs). A backup (emergency) strategy is needed to steer a AV equipped with SBW to safety if its steering actuator fails completely. Differential drive assisted steering (DDAS), which uses torque vectoring (TV) to steer a vehicle in the absence of a steering actuator, has been proposed as a backup strategy for SBW systems. However, vehicle stability control (VSC) – a required feature in most modern vehicles – also relies on torque vectoring to assist vehicles in maintaining stability. In this paper, it is shown through stability analyses on a linearized vehicle model that VSC enhances the stability of a AV equipped with a fully functional SBW system. Conversely, when combined with DDAS as a backup steering strategy for a failed SBW system, VSC typically deteriorates stability. Thus, the benefits of VSC in ensuring stability cannot be taken for granted for AVs, e.g., by designing it as a decoupled controller added on to DDAS. Instead, VSC and DDAS must be designed in an integrated manner to guarantee stability. To support this finding, simulations are conducted where a nonlinear AV model with either SBW or DDAS – both combined with VSC and implemented via model predictive control – are used to carry out an ISO-standard double lane change test at 80 km/h. Using the SBW system, the AV passes the test, both with VSC decoupled from and integrated with the SBW system. However, in the case of SBW failure, the AV with VSC decoupled from DDAS fails the test, while that with VSC integrated with DDAS completes the test successfully.