Ideal car announced at the 2024 Smart Driving Summer Conference that it will roll out “Nationwide Open to Drive” No Map NOA to all Ideal AD Max users in July and will also release fully automatic AES and all-around low-speed AEB in July.
NOA no longer relies on high-precision maps or prior information, navigation coverage is available nationwide, and with the help of spatiotemporal joint planning capabilities, it brings a smoother detour experience. NOA also has the ability to select routes with ultra-long range navigation, allowing smooth passage even at complex intersections.
NOA fully considers the user’s psychological safety boundary, bringing a tacit and reassuring intelligent driving experience with centimeter-level micro-operation. In addition, the upcoming AES function can achieve fully automatic triggering without relying on human assistance, avoiding more high-risk accident risks. The all-around low-speed AEB further expands the active safety risk scenarios, effectively reducing the high-frequency scraping accidents in low-speed parking scenarios.
The new architecture consists of an end-to-end model, VLM visual language model, and world model. The end-to-end model processes regular driving behaviors efficiently, from sensor input to trajectory output. VLM visual language model can understand complex road conditions, navigation maps, and traffic rules. The autonomous driving system will conduct learning and testing in a virtual environment based on the world model.
In July, NOA will be updated with four new features. Users no longer need prior information to use NOA in cities with navigation coverage nationwide. It can even be used in narrow alleys and rural roads.
Next, based on efficient spatiotemporal joint planning capabilities, vehicles avoid and bypass road obstacles more smoothly. Spatiotemporal joint planning synchronizes planning in horizontal and vertical spaces, continuously predicting the spatial interactions between the ego vehicle and other vehicles, planning all feasible trajectories within future time windows. In complex urban intersections, the route selection ability of mapless NOA has also been significantly improved. Mapless NOA uses BEV visual models fused with navigation matching algorithms to real-time perceive changing roadside, road surface arrow signs, and intersection features, fully integrating lane structures and navigation features, effectively solving the problem of complex intersections that are difficult to structure.
