15th Workshop on
Planning, Perception and Navigation for Intelligent Vehicles

Recently, embodied intelligence has achieved transformative breakthroughs, driven by advances in artificial intelligence algorithms and multimodal foundation models. These models, trained on vast cross-modal datasets (vision, language, 3D geometry), empower autonomous robots and systems with grounded multimodal understanding – enabling contextual environmental reasoning, autonomous decision-making under uncertainty, and execution of multi-stage tasks through dynamic action chaining. Such capabilities significantly enhance robotic generalization across unseen scenarios, accelerating the evolution from task-specific automation to adaptive embodied agents capable of zero-shot capability. As we look to the future, enabling robots to operate reliably in open-world environments remains a significant challenge. In such challenging environments, each robot is expected to learn/adapt their surrounding environment via intelligent sensors (such as optical, vision or acoustic sensors) while maintain world models updated through active exploration. Here, hybridizing foundation models’ semantic reasoning (LLMs, VLMs) with classical planning show promise in comprehensively perception of the physical world.

The workshop will feature prominent speakers, and contributions from the intelligent vehicles and mobile robotics community.
The workshop is happening in-person in Hangzhou, CHN. Additionally we welcome participants to listen and contribute virtually via zoom.

This workshop is intended to identify the challenges associated with the software development of autonomous vehicles and to foster discussion about how current research can address them. We invited papers for submission to the workshop related to the topics of ITS, autonomous vehicles, robotics, embodied AI, motion planning, perception and modeling.

Papers to IEEE/RSJ IROS 2025 can be submitted on the [OpenReview Portal]. The page limit is 6 pages, with up to 2 extra pages (with extra page charge). The page limit includes the references, appendixes etc. All papers must be submitted in PDF (up to 6MB) and must follow the IROS double column format. Information and templates are available [here].