RESUMEN
To control autonomous vehicles (AVs) in urban unsignalized intersections is a challenging problem, especially in a hybrid traffic environment where self-driving vehicles coexist with human driving vehicles. In this study, a coordinated control method with proximal policy optimization (PPO) in Vehicle-Road-Cloud Integration System (VRCIS) is proposed, where this control problem is formulated as a reinforcement learning (RL) problem. In this system, vehicles and everything (V2X) was used to keep communication between vehicles, and vehicle wireless technology can detect vehicles that use vehicles and infrastructure (V2I) wireless communication, thereby achieving a cost-efficient method. Then, the connected and autonomous vehicle (CAV) defined in the VRCIS learned a policy to adapt to human driving vehicles (HDVs) across the intersection safely by reinforcement learning (RL). We have developed a valid, scalable RL framework, which can communicate topologies that may be dynamic traffic. Then, state, action and reward of RL are designed according to urban unsignalized intersection problem. Finally, how to deploy within the RL framework was described, and several experiments with this framework were undertaken to verify the effectiveness of the proposed method.
Asunto(s)
Accidentes de Tránsito , Conducción de Automóvil , Comunicación , Humanos , AprendizajeRESUMEN
In this study, two commercial Zr alloys (Zr702 and Zr-2.5Nb) were subjected to the same ß-quenching treatment (water cooling after annealing at 1000 °C for 10 min). Their microstructural characteristics and hardness before and after the heat treatment were well characterized and compared by electron channel contrast (ECC) imaging, electron backscatter diffraction (EBSD) techniques, and microhardness measurements. Results show that after the ß quenching, prior equiaxed grains in Zr702 are transformed into Widmanstätten plate structures (the average width ~0.8 µm) with many fine precipitates distributed along their boundaries, while the initial dual-phase (α + ß) microstructure in Zr-2.5Nb is fully replaced by fine twinned martensitic plates (the average width ~0.31 µm). Differences in alloying elements (especially Nb) between Zr702 and Zr-2.5Nb are demonstrated to play a key role in determining their phase transformation behaviors during the ß quenching. Analyses on crystallographic orientations show that the Burgers orientation relationship is well obeyed in both the alloys with misorientation angles between α plates essentially focused on ~60°. After ß quenching, the hardnesses of both alloys were increased by ~35%-40%. Quantitative analyses using the Hall-Petch equation suggest that such an increase was mainly attributable to phase transformation-induced grain refinements. Since Nb is able to effectively refine the ß-quenched structures, a higher hardness increment is produced in Zr-2.5Nb than in Zr702.