Analytic Approaches for Obtaining the Communications Requirements of String Stable Platooning

Mobile radio networks promise to enable connected automated driving simultaneously with other automated applications, such as those from Industry 4.0. Stringent communications demands on networks are expected from these interconnected automated processes. Since it is anticipated that these appliances share network resources, the issue of shortage of radio resources quickly exacerbates with an increasing number of devices. The radio resource bottleneck exhibits itself as a performance degradation of the connected automated processes. For the case of connected automated driving, the consequences of this communications deterioration can be catastrophic, underscoring the importance of fulfilling their communications demands. From a network perspective, the efficient usage of the shared communications resources is likewise significant; the network aims at maximizing the number of connected users without compromising their operation.
Within connected automated driving, platooning is a relevant application given its benefits in safety, traffic and energy efficiency. These advantages only prevail if the platoon remains string-stable for short intervehicle distances. String stability in connected platooning is determined by distinctive interdependences among the control and communications parameters. While simulation-based studies have broadened the intuition on such interdependencies, their rigorous mathematical description has been missing.
The primary aim of this thesis is to formulate a mathematical framework that reveals the interdependences among the communications and control variables of platoons. From the proposed models, communications requirements of string-stable platoons can be obtained analytically in terms of the platoon dynamic properties. These properties are mostly dictated and stated in terms of the control loop parameters of its member vehicles. By studying the dynamic properties of the vehicle models under a well-defined performance metric, the proposed frameworks relate the control parameters with the communications variables.
Results show that communications requirements of string-stable platoons strongly depend on the control loop parameters dictating the dynamic properties of the vehicles. The resulting upper bounds on packet loss probabilities and transmission intervals do not always fall into the stringent domain as commonly theorized, especially when state estimation techniques are employed on the follower cars.
Consequently, the scalability problem in the network is alleviated by exploiting the vehicles' dynamic properties. Furthermore, the heterogeneity of vehicles induce different communications demands within platoon members, providing additional degrees of freedom for the radio resource allocation.
An implication of understanding communications-control interdependencies is encouraging the development of strategies in either control or communications to compensate deficiencies in their respective co-domain. Another ramification of the findings in this work is on establishing a new paradigm for communications engineers interconnecting heterogeneous automated processes: their distinct dynamic properties provide a type of diversity that is exploitable through radio resource management.
While the here derived methodologies explain the interdependencies between the co-domains of communications and control in connected platooning and the presented results apply only for this use-case, the present research and techniques may serve as incentive and tools for expanding the knowledge and obtaining communications requirements of other networked automated processes.