Cortex 2.0: Grounding World Models in Real-World Industrial Deployment
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866911618221211648 |
|---|---|
| author | Aida, Adriana Amer, Walid Bankovic, Katarina Behl, Dhruv Busch, Fabian Bhalla, Annie Duong, Minh Gienger, Florian Godse, Rohan Grachev, Denis Gulde, Ralf Hagensieker, Elisa Hu, Junpeng Joshi, Shivam Knobloch, Tobias Kumar, Likith LaRocque, Damien Lokesh, Keerthana Moured, Omar Nguyen, Khiem Preyss, Christian Sriganesan, Ranjith Singh, Vikram Sponner, Carsten Tong, Anh Tuscher, Dominik Tuscher, Marc Upputuri, Pavan |
| author_facet | Aida, Adriana Amer, Walid Bankovic, Katarina Behl, Dhruv Busch, Fabian Bhalla, Annie Duong, Minh Gienger, Florian Godse, Rohan Grachev, Denis Gulde, Ralf Hagensieker, Elisa Hu, Junpeng Joshi, Shivam Knobloch, Tobias Kumar, Likith LaRocque, Damien Lokesh, Keerthana Moured, Omar Nguyen, Khiem Preyss, Christian Sriganesan, Ranjith Singh, Vikram Sponner, Carsten Tong, Anh Tuscher, Dominik Tuscher, Marc Upputuri, Pavan |
| contents | Industrial robotic manipulation demands reliable long-horizon execution across embodiments, tasks, and changing object distributions. While Vision-Language-Action models have demonstrated strong generalization, they remain fundamentally reactive. By optimizing the next action given the current observation without evaluating potential futures, they are brittle to the compounding failure modes of long-horizon tasks. Cortex 2.0 shifts from reactive control to plan-and-act by generating candidate future trajectories in visual latent space, scoring them for expected success and efficiency, then committing only to the highest-scoring candidate. We evaluate Cortex 2.0 on a single-arm and dual-arm manipulation platform across four tasks of increasing complexity: pick and place, item and trash sorting, screw sorting, and shoebox unpacking. Cortex 2.0 consistently outperforms state-of-the-art Vision-Language-Action baselines, achieving the best results across all tasks. The system remains reliable in unstructured environments characterized by heavy clutter, frequent occlusions, and contact-rich manipulation, where reactive policies fail. These results demonstrate that world-model-based planning can operate reliably in complex industrial environments. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_20246 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Cortex 2.0: Grounding World Models in Real-World Industrial Deployment Aida, Adriana Amer, Walid Bankovic, Katarina Behl, Dhruv Busch, Fabian Bhalla, Annie Duong, Minh Gienger, Florian Godse, Rohan Grachev, Denis Gulde, Ralf Hagensieker, Elisa Hu, Junpeng Joshi, Shivam Knobloch, Tobias Kumar, Likith LaRocque, Damien Lokesh, Keerthana Moured, Omar Nguyen, Khiem Preyss, Christian Sriganesan, Ranjith Singh, Vikram Sponner, Carsten Tong, Anh Tuscher, Dominik Tuscher, Marc Upputuri, Pavan Robotics Artificial Intelligence I.2.9; I.2.6; I.2.10 Industrial robotic manipulation demands reliable long-horizon execution across embodiments, tasks, and changing object distributions. While Vision-Language-Action models have demonstrated strong generalization, they remain fundamentally reactive. By optimizing the next action given the current observation without evaluating potential futures, they are brittle to the compounding failure modes of long-horizon tasks. Cortex 2.0 shifts from reactive control to plan-and-act by generating candidate future trajectories in visual latent space, scoring them for expected success and efficiency, then committing only to the highest-scoring candidate. We evaluate Cortex 2.0 on a single-arm and dual-arm manipulation platform across four tasks of increasing complexity: pick and place, item and trash sorting, screw sorting, and shoebox unpacking. Cortex 2.0 consistently outperforms state-of-the-art Vision-Language-Action baselines, achieving the best results across all tasks. The system remains reliable in unstructured environments characterized by heavy clutter, frequent occlusions, and contact-rich manipulation, where reactive policies fail. These results demonstrate that world-model-based planning can operate reliably in complex industrial environments. |
| title | Cortex 2.0: Grounding World Models in Real-World Industrial Deployment |
| topic | Robotics Artificial Intelligence I.2.9; I.2.6; I.2.10 |
| url | https://arxiv.org/abs/2604.20246 |