{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,6]],"date-time":"2026-10-06T05:55:00Z","timestamp":1791266100526,"version":"4.1.0"},"reference-count":32,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,10,6]],"date-time":"2026-10-06T00:00:00Z","timestamp":1791244800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Neurorobot."],"abstract":"<jats:sec>\n                    <jats:title>Introduction<\/jats:title>\n                    <jats:p>Bimanual robotic manipulation remains a significant challenge in embodied AI, as it requires synchronized motion planning, inter-arm collision avoidance, and dynamic role allocation over long task horizons. Although existing diffusion-based policies and affordance learning methods have demonstrated considerable capability, they largely treat the environment as static and fail to model how interaction relationships and coordination patterns evolve over time.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>This paper introduces the Predictive Affordance and Coordination Evolution Diffusion Policy (PACE-DP), a unified framework that explicitly forecasts future scene dynamics and models the temporal progression of manipulation affordances and bimanual coordination. PACE-DP integrates three predictive components: (1) a Predictive Scene Module that forecasts latent future object-centric representations; (2) an Affordance and Coordination Evolution Module that constructs temporally evolving interaction graphs and employs a Graph Transformer to predict future relation types, grasp success probabilities, coordination modes, and spatial alignment targets; and (3) a Conditional Diffusion Policy that generates smooth, coordinated action sequences conditioned on fused current and predicted features.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>We evaluate PACE-DP on the RoboTwin 2.0 platform across 10 diverse bimanual tasks, including handover, placement, stacking, and switch operation. Experimental results show that PACE-DP performs favorably compared with baseline methods, achieving higher success rates on coordination-intensive tasks with only marginal computational overhead.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Discussion<\/jats:title>\n                    <jats:p>These findings suggest that explicitly predicting future affordance and coordination evolution provides an effective inductive bias for bimanual manipulation, enabling more reliable coordination without substantially increasing computational cost. PACE-DP thus offers a promising step toward predictive, temporally aware policies for long-horizon bimanual tasks.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/fnbot.2026.1927097","type":"journal-article","created":{"date-parts":[[2026,10,6]],"date-time":"2026-10-06T05:21:47Z","timestamp":1791264107000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Anticipating interactions: a predictive diffusion framework for evolving bimanual manipulation"],"prefix":"10.3389","volume":"20","author":[{"given":"Xukun","family":"Liu","sequence":"first","affiliation":[{"name":"Northwest Institute of Mechanical and Electrical Engineering, Xianyang","place":["Shaanxi, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Fengjuan","family":"Xie","sequence":"additional","affiliation":[{"name":"Northwest Institute of Mechanical and Electrical Engineering, Xianyang","place":["Shaanxi, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xuzhou","family":"He","sequence":"additional","affiliation":[{"name":"School of Energy and Power Engineering, Xi'an Jiaotong University, Xi'an","place":["Shaanxi, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Aifeng","family":"Liu","sequence":"additional","affiliation":[{"name":"Northwest Institute of Mechanical and Electrical Engineering, Xianyang","place":["Shaanxi, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lijun","family":"Yan","sequence":"additional","affiliation":[{"name":"Northwest Institute of Mechanical and Electrical Engineering, Xianyang","place":["Shaanxi, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jianwei","family":"Yang","sequence":"additional","affiliation":[{"name":"Northwest Institute of Mechanical and Electrical Engineering, Xianyang","place":["Shaanxi, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hongsheng","family":"Huang","sequence":"additional","affiliation":[{"name":"Northwest Institute of Mechanical and Electrical Engineering, Xianyang","place":["Shaanxi, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Shenggang","family":"Wei","sequence":"additional","affiliation":[{"name":"Northwest Institute of Mechanical and Electrical Engineering, Xianyang","place":["Shaanxi, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,10,6]]},"reference":[{"key":"B1","article-title":"Is conditional generative modeling all you need for decision-making?","author":"Ajay","year":"2022","journal-title":"arXiv"},{"key":"B2","article-title":"Relational inductive biases, deep learning, and graph networks","author":"Battaglia","year":"2018","journal-title":"arXiv"},{"key":"B3","doi-asserted-by":"publisher","DOI":"10.15607\/RSS.2023.XIX.025","article-title":"Rt-1: robotics transformer for real-world control at scale","author":"Brohan","year":"2022","journal-title":"arXiv"},{"key":"B4","article-title":"Robotwin 2.0: a scalable data generator and benchmark with strong domain randomization for robust bimanual robotic manipulation","author":"Chen","year":"2025","journal-title":"arXiv"},{"key":"B5","doi-asserted-by":"publisher","first-page":"1684","DOI":"10.1177\/02783649241273668","article-title":"Diffusion policy: visuomotor policy learning via action diffusion","volume":"44","author":"Chi","year":"2025","journal-title":"Int. J. Robot. Res."},{"key":"B6","first-page":"465","article-title":"\u201cPILCO: a model-based and data-efficient approach to policy search,\u201d","volume-title":"Proceedings of the 28th International Conference on Machine Learning (ICML-11)","author":"Deisenroth","year":"2011"},{"key":"B7","doi-asserted-by":"crossref","first-page":"7508","DOI":"10.1109\/ICRA57147.2024.10610890","article-title":"\u201cLearning language-conditioned deformable object manipulation with graph dynamics,\u201d","volume-title":"2024 IEEE International Conference on Robotics and Automation (ICRA)","author":"Deng","year":"2024"},{"key":"B8","doi-asserted-by":"publisher","DOI":"10.36227\/techrxiv.175977432.27129012\/v1","article-title":"Embodied AI: from LLMs to world models","author":"Feng","year":"2025","journal-title":"arXiv [Preprint]. arXiv:2509.20021."},{"key":"B9","article-title":"Embodied AI agents: modeling the world","author":"Fung","year":"2025","journal-title":"arXiv"},{"key":"B10","doi-asserted-by":"crossref","first-page":"2864","DOI":"10.1109\/SMC54092.2024.10831380","article-title":"\u201cBi-VLA: vision-language-action model-based system for bimanual robotic dexterous manipulations,\u201d","volume-title":"2024 IEEE International Conference on Systems, Man, and Cybernetics (SMC)","author":"Gbagbe","year":"2024"},{"key":"B11","first-page":"1861","article-title":"\u201cSoft actor-critic: off-policy maximum entropy deep reinforcement learning with a stochastic actor,\u201d","volume-title":"International Conference on Machine Learning","author":"Haarnoja","year":"2018"},{"key":"B12","article-title":"Dream to control: learning behaviors by latent imagination","author":"Hafner","year":"2019","journal-title":"arXiv"},{"key":"B13","article-title":"Mastering Atari with discrete world models","author":"Hafner","year":"2020","journal-title":"arXiv"},{"key":"B14","article-title":"Robomind 2.0: a multimodal, bimanual mobile manipulation dataset for generalizable embodied intelligence","author":"Hou","year":"2025","journal-title":"arXiv"},{"key":"B15","article-title":"Inner monologue: embodied reasoning through planning with language models","author":"Huang","year":"2022","journal-title":"arXiv"},{"key":"B16","article-title":"Planning with diffusion for flexible behavior synthesis","author":"Janner","year":"2022","journal-title":"arXiv"},{"key":"B17","article-title":"Manicm: real-time 3D diffusion policy via consistency model for robotic manipulation","author":"Lu","year":"2024","journal-title":"arXiv"},{"key":"B18","doi-asserted-by":"crossref","first-page":"1796","DOI":"10.1109\/CASE58245.2025.11164107","article-title":"\u201cLearning bimanual manipulation via action chunking and inter-arm coordination with transformers,\u201d","volume-title":"2025 IEEE 21st International Conference on Automation Science and Engineering (CASE)","author":"Motoda","year":"2025"},{"key":"B19","first-page":"3307","article-title":"Data-efficient hierarchical reinforcement learning","volume":"31","author":"Nachum","year":"2018","journal-title":"Adv. Neural Inf. Process. Syst"},{"key":"B20","doi-asserted-by":"crossref","first-page":"7559","DOI":"10.1109\/ICRA.2018.8463189","article-title":"\u201cNeural network dynamics for model-based deep reinforcement learning with model-free fine-tuning,\u201d","volume-title":"2018 IEEE International Conference on Robotics and Automation (ICRA)","author":"Nagabandi","year":"2018"},{"key":"B21","article-title":"Coordinated humanoid manipulation with choice policies","author":"Qi","year":"2025","journal-title":"arXiv"},{"key":"B22","article-title":"A generalist agent","author":"Reed","year":"2022","journal-title":"arXiv"},{"key":"B23","first-page":"77288","article-title":"\u201cDiffusion policy policy optimization,\u201d","volume-title":"International Conference on Learning Representations","author":"Ren","year":"2025"},{"key":"B24","article-title":"Score-based generative modeling through stochastic differential equations","author":"Song","year":"2020","journal-title":"arXiv"},{"key":"B25","article-title":"Graph attention networks","author":"Veli\u010dkovi\u0107","year":"2017","journal-title":"arXiv"},{"key":"B26","article-title":"One-step diffusion policy: fast visuomotor policies via diffusion distillation","author":"Wang","year":"2024","journal-title":"arXiv"},{"key":"B27","article-title":"\u201cCALAMARI: contact-aware and language conditioned spatial Action MApping for contact-RIch manipulation,\u201d","volume-title":"CoRL","author":"Wi","year":"2023"},{"key":"B28","doi-asserted-by":"publisher","first-page":"60966","DOI":"10.52202\/075280-2664","article-title":"Learning environment-aware affordance for 3D articulated object manipulation under occlusions","volume":"36","author":"Wu","year":"2023","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"B29","doi-asserted-by":"crossref","first-page":"2873","DOI":"10.1109\/IROS60139.2025.11246340","article-title":"\u201cGeneralizable humanoid manipulation with 3D diffusion policies,\u201d","volume-title":"2025 IEEE\/RSJ International Conference on Intelligent Robots and Systems (IROS)","author":"Ze","year":"2025"},{"key":"B30","doi-asserted-by":"publisher","author":"Ze","year":"2024","DOI":"10.15607\/RSS.2024.XX.067"},{"key":"B31","doi-asserted-by":"publisher","DOI":"10.15607\/RSS.2024.XX.010","article-title":"Adaptigraph: material-adaptive graph-based neural dynamics for robotic manipulation","author":"Zhang","year":"2024","journal-title":"arXiv"},{"key":"B32","doi-asserted-by":"crossref","first-page":"10838","DOI":"10.1109\/ICRA55743.2025.11128074","article-title":"\u201cScaling diffusion policy in transformer to 1 billion parameters for robotic manipulation,\u201d","volume-title":"2025 IEEE International Conference on Robotics and Automation (ICRA)","author":"Zhu","year":"2025"}],"container-title":["Frontiers in Neurorobotics"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fnbot.2026.1927097\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,10,6]],"date-time":"2026-10-06T05:21:51Z","timestamp":1791264111000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fnbot.2026.1927097\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,10,6]]},"references-count":32,"alternative-id":["10.3389\/fnbot.2026.1927097"],"URL":"https:\/\/doi.org\/10.3389\/fnbot.2026.1927097","relation":{},"ISSN":["1662-5218"],"issn-type":[{"value":"1662-5218","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,10,6]]},"article-number":"1927097"}}