July 15th,2025New energyForum on the development of new quality of intelligent vehiclesIt was successfully held in Changchun.
The conference was held to“New quality leads the future intelligently”As the theme, leaders of relevant government departments, experts from Tsinghua University, Harbin University of Technology, Jilin University, etc., as well as representatives of Geely, Lantu Automobile, Jiangqi Group, FAW Group, Great Wall Motor, Neusoft Rich, Valeo, Horizon, SemiDrive Technology, JD Technology, Lingchong, AAC Technology, ZTE Microelectronics, Geometry Partners, Aodong New Energy, Dongchi New Energy and other vehicle and supply chain enterprises attended the meeting.
Zhang Chunda, Vice Mayor of Changchun Municipal People’s Government, and Zhang Guigang, Deputy Director of Jilin Provincial Department of Commerce, ChinaElectric vehicleZhang Yongwei, vice chairman and secretary-general of the 100 People’s Association, Chen Dezhi, president of Changchun Council for the Promotion of International Trade, and Han Baoshuang, chairman of Barry International Convention and Exhibition Group, attended the meeting.
Zhang Chunda, vice mayor of Changchun Municipal People’s GovernmentatIn his speech, he said that the current global scientific and technological revolution and industrial transformation are surging.New energy vehiclesWith rapid development, the world’s automobile pattern is being deeply adjusted and reshaped, and only open cooperation and collaborative innovation can win the future. Changchun City is unswervingly following the road of innovation-driven development, setting up a special fund of 10 billion yuan for the automobile industry, introducing support policies covering the entire industry chain, and continuously seeking breakthroughs in key areas such as “three electrics”, intelligent cockpit, and autonomous driving with FAW as the core.
Zhang Yongwei, vice chairman and secretary-general of the China Electric Vehicle 100 AssociationHe said that China has achieved electrified lane change and overtaking, and the terminal market is the first in terms of intelligence, and the industrial ecology is becoming increasingly rich. The current core tasks include two stage goals: from 2025 to 2030, accelerate the popularization of assisted driving, expand the user base, and cultivate an intelligent driving culture. At the same time, he stressed that our goal should not only focus on the popularization of low-end technology, but also quickly put forward the development goals of L3 and higher levels of autonomous driving, and countries and enterprises that take the lead in realizing the application of L3 and above technologies will have a key advantage. Combined with the dynamic judgment of enterprises, 2030 and the next two or three years may be the window period for L3 and L4 to move from pilot to large-scale application.
The main forum was divided into two sessions in the morning and afternoon, respectively by Shi Jianhua, deputy secretary-general of the China Electric Vehicle 100 Association and president of the Chebai Think Tank Research Institute, and the China Electric Vehicle 100 AssociationIntelligent networkingZhao Zesheng, a distinguished expert of the institute, presided over the meeting.
01
Expert insights unlock technological breakthroughs and industrial directions
At the meeting, expert representatives shared their innovative ideas in key areas of smart cars and put forward new ideas for industrial development.
Wang Jianqiang, Dean and Professor of the School of Vehicles and Transportation of Tsinghua UniversityIt is pointed out that the current intelligent car safety technology is mainly promoted along two paths: first, it is rule-driven, with the advantages of clear logic and strong interpretability, but it faces bottlenecks such as weak generalization ability and difficulty in covering long-tail scenarios; The second is data-driven, with independent learning ability and good generalization, but there are problems such as “black box” attributes, difficult traceability, and insufficient ability to identify extreme scenes. Both paths have their own advantages and disadvantages, and it is difficult to support the evolution of intelligent driving to a higher level alone. In order to break through the bottleneck of key technologies, he proposed a “cognitive-driven” route, which integrates human brain-like cognition and knowledge experience to inspire people, and opens up the integration channel of rules and data through the three major links of perception-cognition-decision-making to achieve accurate perception, deep cognition and safety decision-making of the system in complex traffic environments.
Cai Wei, tenured honorary professor of Harbin University of Technology, national outstanding engineer, and foreign academician of the Russian Academy of EngineeringIt is mentioned that as the charging system voltage moves towards 1000V or even 1500V, wide bandgap power devices such as silicon carbide have gradually become the mainstream, but they also bring problems such as thermal runaway and high system costs. In terms of motor technology route, the current mainstream is permanent magnet synchronous motor, and Tesla has also shifted from induction motor to permanent magnet solution. He emphasized that electrification + cleanliness is a long-term national strategic goal, and the electric drive system must be independent and controllable as a core component. Promoting the development of the “four modernizations”, namely electrification, intelligence, low-carbonization, and globalization, is the direction of the industry’s joint efforts.
Zhang Yongwei, vice chairman and secretary-general of the China Electric Vehicle 100 AssociationIt said: In the era of intelligence, the focus of enterprise competitiveness has shifted to intelligence and AI capabilities. In the future, the proportion of machinery in automobile costs will drop from 70% to less than 30%, and the proportion of electronics and software will reach 70%, and intelligence and cost will become the primary factors in car purchase decisions. He believes that the development of intelligent enterprises can be divided into three stages: using AI to assist R&D and management; producing intelligent products, such as AI cars; Transform into an AI underlying enterprise and create a series of AI products.
Lu Dang, professor of Jilin UniversityHe said that intelligent chassis is one of the key technologies for intelligent car control, and intelligent tires and their domain control software will become an important foundation for software-defined vehicles in the future. As the only component of the car in contact with the road, all the driver’s control commands generate tire force through the interaction between the tire and the ground, which in turn changes the car movement state, so the tire is the core of the chassis, and the tire force is the root cause of the car movement.
02
cartEnterprise theory, multiple technical routes anchor intelligence
Representatives of vehicle companies shared their practices in technical route selection, brand upgrading and ecological construction, showing the diversified development path of new energy intelligent vehicles.
Lu Fang, CEO of Lantu Automobile Technology Co., LtdIt is believed that in the context of the current scientific and technological revolution and the acceleration of industrial transformation, new products, new technologies and new formats are emerging, and the competitive landscape is becoming increasingly complex. How to break through low-level competition and promote high-quality development in China’s automobile industry is a topic that all enterprises must face. We should adhere to technological innovation and achieve a leap from “competing parameters” to “creating value”; adhere to management innovation to achieve a leap from transaction relationship to partnership; adhere to brand innovation to achieve a leap from single consumption to long-term trust; Adhere to open innovation and achieve a leap from Chinese products to Chinese brands.
Ma Cuibing, director of Jiangqi Group Holding Company and deputy general manager of Jiangqi Group Co., LtdHe said that the automobile industry is in a critical stage of development of multi-technological innovation and multi-industry integration and development. The door of traditional cars is closing at an accelerated pace, but new cars, new ecology, and new value chains are also accelerating, and Chinese car companies have ushered in an excellent opportunity to enter the high-end. JAC has been with it since 2019HuaweiSince reaching a comprehensive strategic cooperation, the two sides have deeply integrated the whole value chain link such as R&D, supply chain, manufacturing, and sales service. The Zunjie brand is a typical practice: JAC and Huawei have set up an exclusive R&D and delivery team of more than 5,000 people, and jointly built the Zunjie Gigafactory. In addition, JAC has joined forces with 227 top suppliers to build a world-class supply chain ecology and promote breakthroughs in the field of ultra-luxury intelligent new energy.
Zhou Shiying, vice president of FAW R&D Institute and CEO of the Jiuzhang platformIt is proposed that with the rapid evolution of AI large models, in-vehicle applications have shifted from traditional software-defined vehicles to “AI-defined vehicles”. Since 2024, the cockpit and intelligent driving system have almost been connected to AI models by default, but they pose great challenges to the underlying chip, OS architecture and application layer compatibility. At present, most OEMs have not yet fully separated software and hardware. In addition, in the context of tightening capital, more and more enterprises are turning to embodied intelligence. As an early form of “humanoid robot”, smart cars have natural advantages in embedded systems, neural networks and AI connection foundations.
Pang Shuyi, deputy chief engineer of Great Wall Motor’s intelligent drivingIt is proposed that as the earliest OEM to lay out end-to-end technology, Great Wall has anchored the “end-to-end + no picture” technical solution from the early stage of development and launched the end-to-end combined driving assistance model SEE. The model adopts the “data-driven + model security + security policy bottom-up” model, and on the basis of the end-to-end large model, security arbitration is carried out by adding security assessment networks and manual rules to ensure system security; At the same time, the deep integration of VLA models further improves the semantic understanding and coupling of complex road conditions, generates more anthropomorphic driving strategies, and optimizes driving experience.
03
Supply chain co-exploration, new quality reconstruction and ecological collaboration
As an important part of the industrial ecosystem, supply chain enterprises also shared their thoughts on technology empowerment and ecological collaboration.
Du Qiang, President and CTO of Neusoft Automotive Technology (Shanghai) Co., LtdHe said that in the past six months, AI has had a significant impact on the entire automotive industry, automotive experience and the core competitiveness of automobiles. In response to the question of “whether AI subverts or reconstructs traditional vehicle OS”, he believes that AI is an incremental construction of automotive systems, not a replacement. China’s software supply chain has grown and matured in the field of underlying operating systems, and has the ability to quickly adapt chips and electronic and electrical architectures to meet the high-speed iteration needs of car companies. The efficiency of software development empowered by AI technology has been greatly improved, and toolchains and semi-automated development have helped lower the threshold and improve human efficiency.
Gu Jianmin, Chief Technology Officer of Valeo ChinaIt is believed that the current automotive industry is facing multiple key trends, in addition to new energy electrification, intelligent driving, and intelligent cockpits, in-vehicle infotainment systems and personalized visual elements (such as lights) are also becoming more and more important. Lamps are not only lighting tools, but also important carriers of design and information interaction. Behind all these technologies, software is becoming a key force in defining the car.
Lu Peng, Vice President of Horizon, Head of Strategy Department & Intelligent Driving Product Planning and Marketing DepartmentIt is believed that intelligent driving is similar to a mobile phone base station, with complex technology but difficult to form differentiation in the end, and the core is “rapid iteration”. However, he reminded that in the industrial cycle, it is necessary to balance cost reduction and technology iteration, the industry is not yet mature, excessive cost reduction will sacrifice user experience and safety, and experience upgrades should be promoted through technological iteration (such as computing power improvement).
Zhang Xitong, general manager of SemiDrive’s MCU product lineHe said that the scale of the automobile market continues to expand, and the localization rate of intelligent cockpits and vehicle control chips has become the focus. In recent years, car companies have actively and spontaneously adopted domestic chips to significantly accelerate the localization process. Supply chain relationships are being reconstructed, and more and more local chip companies are involved in the early stages of vehicle design, jointly defining and developing products with OEMs and Tier 1 to meet rapid iteration and high-quality requirements.
Zhou Qiang, CEO of Lingchong New Energy Technology Co., LtdIt is believed that vehicle-network interaction is conducive to alleviating the pressure on the power grid brought about by the scale of new energy vehicles and ensuring the stability of the power grid. help the consumption of renewable energy and promote the low-carbon transformation of the energy structure; Reduce the cost of using new energy vehicles, accelerate market popularization, promote the upgrading of the energy storage industry, improve the new energy industry chain to support the construction of the energy Internet, and help achieve the “double carbon” goal. At present, vehicle-network interaction is still in the early stage of commercialization, but it is accelerating its development towards scenario-based, systematic and platform-based. The park-level energy system will become a key scenario for the implementation of vehicle-network interaction.
Sun Xuejing, chief scientist of AAC TechnologyHe said that with the rapid evolution of intelligent cockpits, vehicle acoustics are changing from “auditory systems” to the core technical support for building a “third space”. The acoustic system is no longer just “audible”, but is related to the underlying support of human-vehicle interaction experience, active safety and even ecological collaboration
Zhang Xianhong, chief technology officer of Shanghai Geometry Partner Intelligent Driving Co., LtdIt is believed that when all-day and all-weather safe intelligent driving moves from luxury car privileges to national standards, new quality productivity can realize the inclusiveness of high-end technology. The software and hardware integrated solution of the “vision + 4D imaging radar” heterogeneous information fusion perception technology path will continuously improve the safety performance of intelligent assisted driving systems, allow more affordable passenger cars and commercial vehicles to enjoy a more secure and stable driving experience, and accelerate the arrival of the era of inclusive safety in China’s intelligent driving.
2025 New Energy Intelligent Vehicle New Quality Development ForumIt is not only a feast of ideas, but also shows the innovative vitality and multiple paths of China’s new energy intelligent vehicle industry, providing new ideas for promoting technological collaborative innovation and strengthening ecological integration.
The dynamics of this forum and the views of the guests will be released through the official channels of the China Electric Vehicle 100 Association, so stay tuned!
