{"id":{"repo_id":"mit","oai_identifier":"oai:dspace.mit.edu:1721.1/156571"},"canonical_url":"https://search.dev.ndltd.org/etd/mit/oai:dspace.mit.edu:1721.1/156571","repository":{"repo_id":"mit","name":"MIT","base_url":"https://dspace.mit.edu/oai/request"},"display":{"title":"Robust Scene and Object Generalization of Neural Policies Trained in Synthetic Environments","abstract":"Achieving generalization for autonomous robotic systems operating in real-world environments remains a significant challenge. Training robots solely in simulations can be limiting due to the \"sim-to-real gap\"– discrepancies between simulated and real-world conditions. We present two novel approaches to enhance the generalization capabilities of autonomous quadrotor navigation systems when transferring from simulation to the real world. Our f irst approach integrates a 3D Gaussian Splatting radiance field with a quadrotor flight dynamics engine to generate high-quality, photorealistic training data. We design imitation learning schemes to train liquid time-constant neural networks on this data. Through rigorous evaluations, we demonstrate successful zero-shot transfer of the learned navigation policies from simulation to real-world flight, exhibiting generalization to complex, multi-step tasks in novel indoor and outdoor environments. Notably, we showcase autonomous quadrotor policies trained entirely in simulation that can be directly deployed in the real world without fine-tuning. Our method leverages the complementary strengths of photorealistic rendering and irregularly time-sampled data augmentation for enhancing generalization with liquid neural networks. Additionally, we compose off-the-shelf vision-and-language models with neural policies, enabling real-world generalization to complex objects and instructions unseen during training. To the best of our knowledge, this is the first report of zero-shot sim-to-real transfer and semantic generalization for autonomous quadrotor navigation using imitation learning. Our key contributions include: (1) a dynamics-augmented Gaussian splatting simulator, (2) implicit closed-loop augmentation via expert trajectory design, (3) robustifying liquid neural networks through irregularly sampled data, (4) extensive simulation and real-world validation, (5) demonstrating zero-shot real-world transfer capabilities, and (6) enabling zero-shot instruction generalization to novel objects using multimodal representations.","abstract_html":"Achieving generalization for autonomous robotic systems operating in real-world environments remains a significant challenge. Training robots solely in simulations can be limiting due to the "sim-to-real gap"– discrepancies between simulated and real-world conditions. We present two novel approaches to enhance the generalization capabilities of autonomous quadrotor navigation systems when transferring from simulation to the real world. Our f irst approach integrates a 3D Gaussian Splatting radiance field with a quadrotor flight dynamics engine to generate high-quality, photorealistic training data. We design imitation learning schemes to train liquid time-constant neural networks on this data. Through rigorous evaluations, we demonstrate successful zero-shot transfer of the learned navigation policies from simulation to real-world flight, exhibiting generalization to complex, multi-step tasks in novel indoor and outdoor environments. Notably, we showcase autonomous quadrotor policies trained entirely in simulation that can be directly deployed in the real world without fine-tuning. Our method leverages the complementary strengths of photorealistic rendering and irregularly time-sampled data augmentation for enhancing generalization with liquid neural networks. Additionally, we compose off-the-shelf vision-and-language models with neural policies, enabling real-world generalization to complex objects and instructions unseen during training. To the best of our knowledge, this is the first report of zero-shot sim-to-real transfer and semantic generalization for autonomous quadrotor navigation using imitation learning. Our key contributions include: (1) a dynamics-augmented Gaussian splatting simulator, (2) implicit closed-loop augmentation via expert trajectory design, (3) robustifying liquid neural networks through irregularly sampled data, (4) extensive simulation and real-world validation, (5) demonstrating zero-shot real-world transfer capabilities, and (6) enabling zero-shot instruction generalization to novel objects using multimodal representations.","abstract_has_math":false,"creators":["Quach, Alex H."],"institution":"Massachusetts Institute of Technology","degree_name":"Master","degree_level":null,"degree_discipline":null,"degree_department":"Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science","school":null,"contributors":[],"advisors":["Rus, Daniela"],"committee_chairs":[],"committee_members":[],"year":2024,"date_issued":"2024-05","date_published":"2024-05","updated_at":"2026-07-22T22:20:53Z","subjects":[],"languages":[],"rights":["Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0)","Copyright retained by author(s)"],"rights_urls":["https://creativecommons.org/licenses/by-nc-nd/4.0/"],"identifier_entries":[]},"links":{"outbound_url":"https://hdl.handle.net/1721.1/156571","outbound_label":"Handle","outbound_source":"dc:identifier.uri"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.advisor","label":"Advisor","values":["Rus, Daniela"]},{"key":"dc:contributor.department","label":"Department","values":["Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science"]},{"key":"dc:creator","label":"Author","values":["Quach, Alex H."]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.accessioned","label":"Dc Date Accessioned","values":["2024-09-03T21:08:18Z"]},{"key":"dc:date.available","label":"Dc Date Available","values":["2024-09-03T21:08:18Z"]},{"key":"dc:date.issued","label":"Date","values":["2024-05"]},{"key":"dc:publisher","label":"Institution","values":["Massachusetts Institute of Technology"]},{"key":"dc:type","label":"Dc Type","values":["Thesis"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Master","Master of Engineering in Electrical Engineering and Computer Science"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:rights","label":"Dc Rights","values":["Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0)","Copyright retained by author(s)"]},{"key":"dc:rights.uri","label":"Rights URI","values":["https://creativecommons.org/licenses/by-nc-nd/4.0/"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://hdl.handle.net/1721.1/156571"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["Achieving generalization for autonomous robotic systems operating in real-world environments remains a significant challenge. Training robots solely in simulations can be limiting due to the \"sim-to-real gap\"– discrepancies between simulated and real-world conditions. We present two novel approaches to enhance the generalization capabilities of autonomous quadrotor navigation systems when transferring from simulation to the real world. Our f irst approach integrates a 3D Gaussian Splatting radiance field with a quadrotor flight dynamics engine to generate high-quality, photorealistic training data. We design imitation learning schemes to train liquid time-constant neural networks on this data. Through rigorous evaluations, we demonstrate successful zero-shot transfer of the learned navigation policies from simulation to real-world flight, exhibiting generalization to complex, multi-step tasks in novel indoor and outdoor environments. Notably, we showcase autonomous quadrotor policies trained entirely in simulation that can be directly deployed in the real world without fine-tuning. Our method leverages the complementary strengths of photorealistic rendering and irregularly time-sampled data augmentation for enhancing generalization with liquid neural networks. Additionally, we compose off-the-shelf vision-and-language models with neural policies, enabling real-world generalization to complex objects and instructions unseen during training. To the best of our knowledge, this is the first report of zero-shot sim-to-real transfer and semantic generalization for autonomous quadrotor navigation using imitation learning. Our key contributions include: (1) a dynamics-augmented Gaussian splatting simulator, (2) implicit closed-loop augmentation via expert trajectory design, (3) robustifying liquid neural networks through irregularly sampled data, (4) extensive simulation and real-world validation, (5) demonstrating zero-shot real-world transfer capabilities, and (6) enabling zero-shot instruction generalization to novel objects using multimodal representations."]},{"key":"dc:description.degree","label":"Dc Description Degree","values":["M.Eng."]},{"key":"dc:title","label":"Title","values":["Robust Scene and Object Generalization of Neural Policies Trained in Synthetic Environments"]}]}],"canonical_facts":{"dc:contributor.advisor":["Rus, Daniela"],"dc:contributor.department":["Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science"],"dc:creator":["Quach, Alex H."],"dc:date.accessioned":["2024-09-03T21:08:18Z"],"dc:date.available":["2024-09-03T21:08:18Z"],"dc:date.issued":["2024-05"],"dc:description.abstract":["Achieving generalization for autonomous robotic systems operating in real-world environments remains a significant challenge. Training robots solely in simulations can be limiting due to the \"sim-to-real gap\"– discrepancies between simulated and real-world conditions. We present two novel approaches to enhance the generalization capabilities of autonomous quadrotor navigation systems when transferring from simulation to the real world. Our f irst approach integrates a 3D Gaussian Splatting radiance field with a quadrotor flight dynamics engine to generate high-quality, photorealistic training data. We design imitation learning schemes to train liquid time-constant neural networks on this data. Through rigorous evaluations, we demonstrate successful zero-shot transfer of the learned navigation policies from simulation to real-world flight, exhibiting generalization to complex, multi-step tasks in novel indoor and outdoor environments. Notably, we showcase autonomous quadrotor policies trained entirely in simulation that can be directly deployed in the real world without fine-tuning. Our method leverages the complementary strengths of photorealistic rendering and irregularly time-sampled data augmentation for enhancing generalization with liquid neural networks. Additionally, we compose off-the-shelf vision-and-language models with neural policies, enabling real-world generalization to complex objects and instructions unseen during training. To the best of our knowledge, this is the first report of zero-shot sim-to-real transfer and semantic generalization for autonomous quadrotor navigation using imitation learning. Our key contributions include: (1) a dynamics-augmented Gaussian splatting simulator, (2) implicit closed-loop augmentation via expert trajectory design, (3) robustifying liquid neural networks through irregularly sampled data, (4) extensive simulation and real-world validation, (5) demonstrating zero-shot real-world transfer capabilities, and (6) enabling zero-shot instruction generalization to novel objects using multimodal representations."],"dc:description.degree":["M.Eng."],"dc:identifier.uri":["https://hdl.handle.net/1721.1/156571"],"dc:publisher":["Massachusetts Institute of Technology"],"dc:rights":["Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0)","Copyright retained by author(s)"],"dc:rights.uri":["https://creativecommons.org/licenses/by-nc-nd/4.0/"],"dc:title":["Robust Scene and Object Generalization of Neural Policies Trained in Synthetic Environments"],"dc:type":["Thesis"],"thesis:degree_name":["Master","Master of Engineering in Electrical Engineering and Computer Science"]},"updated_at":"2026-07-22T22:20:53Z"}