Effective modeling of how human travelers learn and adjust their travel behavior from interacting with transportation systems is critical for system assessment and planning. However, this task is also difficult due to the complex cognition and decision-making involved in such behavior. Recent research has begun to leverage Large Language Model (LLM) agents for this task. Building on this, we introduce a novel dual-agent framework that enables continuous learning and alignment between LLM agents and human travelers on learning and adaptation behavior from online data streams. Our approach involves a set of LLM traveler agents, equipped with a memory system and a learnable persona, which serve as simulators for human travelers. To ensure behavioral alignment, we introduce an LLM calibration agent that leverages the reasoning and analytical capabilities of LLMs to train the personas of these traveler agents. Working together, this dual-agent system is designed to track and align the underlying decision-making mechanisms of travelers and produce realistic, adaptive simulations. Using a real-world dataset from a day-to-day route choice experiment, we show our approach significantly outperforms existing LLM-based methods in both individual behavioral alignment and aggregate simulation accuracy. Furthermore, we demonstrate that our method moves beyond simple behavioral mimicry to capture the evolution of underlying learning processes, a deeper alignment that fosters robust generalization. Overall, our framework provides a new approach for creating adaptive and behaviorally realistic agents to simulate travelers' learning and adaptation that can benefit transportation simulation and policy analysis.

The transport ministry will launch priority lanes for autonomous vehicles on a roughly 100-kilometer section of the Shin-Tomei Expressway to conduct test runs of self-driving trucks, starting on March 3. The trial is expected to last about one year. The move is part of a plan to create a road network that can handle so-called Level 4 autonomous driving, or unmanned driving under certain conditions, amid labor shortages in the logistics industry. The lanes will run from the Surugawan-Numazu rest area to the Hamamatsu rest area, both in Shizuoka Prefecture. One of the three lanes in both the eastbound and westbound directions of the expressway will be designated priority lanes between 10 p.m. on weekdays and 5 a.m. the following morning.

The development of safety-oriented research and applications requires fine-grain vehicle trajectories that not only have high accuracy, but also capture substantial safety-critical events. However, it would be challenging to satisfy both these requirements using the available vehicle trajectory datasets do not have the capacity to satisfy both.This paper introduces the CitySim dataset that has the core objective of facilitating safety-oriented research and applications. CitySim has vehicle trajectories extracted from 1140 minutes of drone videos recorded at 12 locations. It covers a variety of road geometries including freeway basic segments, signalized intersections, stop-controlled intersections, and control-free intersections. CitySim was generated through a five-step procedure that ensured trajectory accuracy. The five-step procedure included video stabilization, object filtering, multi-video stitching, object detection and tracking, and enhanced error filtering. Furthermore, CitySim provides the rotated bounding box information of a vehicle, which was demonstrated to improve safety evaluations. Compared with other video-based critical events, including cut-in, merge, and diverge events, which were validated by distributions of both minimum time-to-collision and minimum post-encroachment time. In addition, CitySim had the capability to facilitate digital-twin-related research by providing relevant assets, such as the recording locations' three-dimensional base maps and signal timings.

Subaru Corp. aims to start selling vehicles equipped with the equivalent of "level-2" autonomous technology, which can steer, accelerate and slow down on ordinary roads, in the second half of the 2020s, company officials said Tuesday. But developing such vehicles to run on ordinary roads has been a challenge, due to the need to respond to unexpected scenarios such as pedestrians' movements in order to avoid accidents. Subaru plans to develop a next-generation system using its EyeSight Driver Assist Technology and artificial intelligence to recognize a traffic lane even when the white line on the roads cannot be seen. Level-2 technologies allow drivers to take their hands off the steering wheel under certain conditions, but drivers are still required to monitor the vehicle's driving at all times. Earlier this year, Toyota launched new models of its luxury sedan Lexus LS and hydrogen-powered Mirai that are equipped with level-2 assistant technologies.

Driverless cars will soon hit expressways for testing in Chinese capital Beijing, as the municipal authority has approved the gradual opening of several expressway sections for unmanned vehicles. A 10-km expressway between Beijing's fifth and sixth ring roads will be opened for such testings first. Six other expressway sections will be opened later to add another 143-km stretch to the city's high-speed testing roads for autonomous vehicles, according to a work group overseeing the city's innovation for mobility intelligence. Jiang Guangzhi, deputy director of the Beijing Municipal Bureau of Economy and Information Technology, said that the high-speed test of self-driving vehicles has a high entry threshold. Testing vehicles should be equipped with relevant devices and connected to a cloud computing platform for real-time data transmission to ensure the safety of expressway traffic.

Honda Motor Co. said Thursday it will start offering from Friday the revamped Legend sedan in Japan equipped with "level-3" autonomous technology as the auto industry faces intensifying competition to develop driverless vehicles and a collision-free society. It is the world's first vehicle to hit the market that allows the driver to engage in different tasks such as reading and watching TV when the car is in certain conditions such as congested traffic on expressways, the Japanese transport ministry said. But in the case of an emergency the driver needs to take full control of the vehicle. "Autonomous technology has the potential to reduce the driver's burden while eliminating human errors that cause traffic accidents," Yoichi Sugimoto, executive chief engineer of Honda R&D Co., said in an online press conference. Honda plans to offer 100 units domestically for a suggested retail price of ¥11 million ($103,000) that will only be available on a three-year lease.

KUALA LUMPUR: Highways operator Prolintas has achieved another milestone by developing and integrating artificial intelligence (AI) and machine learning as part of their highway management. The highway concessionaire has been using the Smart Surveillance System (S3) and Prolintas Integrated Maintenance Escalations (Prime) which are based on smart technology system. This system upgrade has put Prolintas as the first in the country to utilise such advanced technology system in highway management operations. Prolintas group chief executive officer Datuk Mohammad Azlan Abdullah said the system upgrade was aimed to improve the level of highway efficiency. "By using S3, we will be able to do highway surface monitoring, faulty assets detection, road accidents, foreign object, stagnant water areas and animal presence. "Apart from that, it will also be able to do live video streaming (of the highway), real-time integrated notification and reporting.

Honda Motor Co. is preparing this year to become the first Japanese automaker to launch a vehicle with "level 3" autonomous driving, according to sources familiar with the matter. Level 3 autonomy frees up the driver to engage in different activities, such as reading or watching TV. Honda's car is expected to offer such hands-off capabilities only in slow traffic on congested expressways, the sources said. Audi AG already sells cars capable of level 3 autonomy in Japan, with self-driving available at busy times on expressways. Honda's car will likely be introduced after Japanese carmakers jointly test autonomous vehicle operations on public roads in Tokyo around July, they said.

Honda Motor Co. will be the first automaker in Japan to release a domestically produced car with "level three" autonomy, sources have said. Level three autonomous driving technologies will be mounted on Honda's new Legend luxury sedan, set to be launched as early as summer 2020, the sources said. With the level three technologies, drivers will be able to leave operation of their car to the vehicle itself during traffic congestion on expressways. Next spring, Japan will enforce a revised road traffic law to permit the driving of level three autonomous vehicles on public roads. With the technologies that Honda will mount on the Legend, drivers will not need to operate the accelerator or the wheel during congestion on expressways.

The Cabinet on Friday approved rules for operating partially self-driving vehicles, paving the way for the use of autonomous vehicles on public roads. Autonomous driving technology is classified into five levels, ranging from Level 1, which allows either steering, acceleration or braking to be automated, to fully automated Level 5. The government plans to enforce an ordinance defining violations and setting penalties by May next year as it envisions the use of Level 3 vehicles, which allow conditionally automated driving, on expressways in 2020. The newly decided penalties apply to the inappropriate use of Level 3 autonomous driving technologies, which require users to switch to manual operations when preset conditions regarding road type, driving speed, weather, time of day and other factors are no longer met. Violators of the ordinance will face fines of up to ¥12,000 ($110) depending on vehicle size.