Optimal internal boundary control of lane-free automated vehicle traffic

Linear–Quadratic regulators for internal boundary control of lane-free automated vehicle traffic

“…influence the movement of vehicles in front of them. In this context, the internal boundary control concept, introduced by Malekzadeh et al (2020), exploits the lane-free principle of TrafficFluid, specifically the property that the road capacity may exhibit incremental (increasing or decreasing) changes in response to corresponding incremental (widening or narrowing) changes of the road width. This is in contrast to lane-based roads and traffic, where capacity changes may only occur if the road width is changed by a lane or a lane-multiple.…”

“…The use of tidal flow control systems in lane-based traffic is not widespread for a number of reasons (see Malekzadeh et al (2020)), such as: harsh resolution of infrastructure sharing (only by lane quanta) among the two traffic directions; serious counter-problems due to frequent merging or diverging traffic at lane-drop or lane-gain areas; safety-induced time-delays after each lane switch.…”

“…In contrast, in a lane-free CAV traffic environment, the mentioned difficulties are largely mitigated (Malekzadeh et al (2020)). More specifically:…”

“…Even for infrastructures experiencing strong demand in both directions quasi-simultaneously, real-time internal boundary control may intensify the road utilization and lead to sensible improvements. Malekzadeh et al (2020) analyzed the internal boundary control problem and demonstrated its high improvement potential by formulating and solving an open-loop nonlinear constrained optimal control problem, in the form of a convex Quadratic Programming (QP) problem. That approach may be used in the frame of MPC (Model Predictive Control) with updated external demand predictions for real-time application; however, simpler real-time algorithms, that do not call for demand prediction and reach a similar efficiency level as the mentioned QP approach, are preferable.…”

“…The well-known CTM is used for controller design as well as simulation testing. Using the same highway stretch as in (Malekzadeh et al, 2020), a series of demand scenarios are considered, and the performance of the LQ regulators is compared to each other, to the no-control case, as well as to the optimal results obtained using the QP formulation by Malekzadeh et al (2020) with perfect prediction.…”