"Pioneer" on the competition field.

The 14th China Intelligent Vehicle Future Challenge was held in Changshu, Jiangsu province, from Nov 22 to 23, featuring 76 teams from universities, enterprises, and research institutes. Xi'an Jiaotong University (XJTU)'s unmanned intelligent vehicle, "Pioneer", won first place in the real traffic environment challenge, marking its fifth consecutive championship.

The real traffic environment challenge aims to address key issues faced by fully autonomous driving technology in open, dynamic environments. It focuses on testing vehicles' autonomous driving capabilities under extreme conditions such as adverse weather, unexpected obstacles, uncooperative targets, and road construction in natural traffic scenarios.

It also assesses vehicles' human-machine interaction capabilities and autonomous driving technology in complex traffic scenarios, emphasizing intelligent decision-making in real-world traffic environments.

This year's challenge simulated a common scenario in which autonomous vehicles navigate from public roads to unmapped, enclosed residential or commercial parking areas located at the end of their route. Participants were encouraged to overcome current technological limitations by using innovations like large language models, advanced computer vision models, and general AI agents, aiming to achieve refined, personalized, and seamless autonomous driving services.

"Pioneer" was developed by XJTU's Institute of Artificial Intelligence and Robotics. Established in 1986, the institute is one of China's first dedicated research institutions focusing on artificial intelligence. Since 2002, the institute's research team has been at the forefront of key technology research and system validation for autonomous intelligent vehicles.