EIDA: Execution-Interface Dynamics Adaptation for Real-to-Sim-to-Real Robot Navigation

Yiwei Qian; Shanze Wang; Qingyuan Hu; Xinming Zhang; Wei Zhang

cs.ROOct 1, 2026

EIDA: Execution-Interface Dynamics Adaptation for Real-to-Sim-to-Real Robot Navigation

Authors: Yiwei Qian, Shanze Wang, Qingyuan Hu, Xinming Zhang, Wei Zhang

Organizations: Eastern Institute of Technology, Ningbo, China · National University of Singapore, Singapore · Department of Aeronautical and Aviation Engineering, The Hong Kong Polytechnic University, Hong Kong · University of Science and Technology of China, Hefei, China

Abstract

Simulation-to-robot transfer can fail when velocity commands produce motion and feedback that differ from those modeled during policy training. We present execution-interface dynamics adaptation (EIDA), which fits these responses from target-platform execution data without reconstructing actuator dynamics. A model of body-frame pose increments updates simulator geometry, while a separate model predicts the velocity feedback observed by the policy; a short history of velocity feedback is included in the policy input. The fitted models are used within a lightweight GPU-parallel simulator. On the full Jackal and Go2 validation sets, the fitted models reduced position and yaw prediction errors relative to the simulator's predefined motion model. Across 100 benchmark navigation environments evaluated in a separate physics-based simulator, EIDA achieved the highest success rate and navigation score among the compared learned policies, both with and without global guidance. Feedback ablations further supported the need to match policy-facing velocity estimates. On a physical Unitree Go2, EIDA reached the goal without collision in all 20 static-scene trials, compared with 4 of 20 for the baseline. These results show that execution-interface adaptation can improve navigation transfer without detailed actuator simulation.

Figures & tables

Fig. 1: Sim-to-real execution mismatch can cause navigation failures. EIDA incorporates models learned from real executions into lightweight simulation to improve navigation policy transfer to physical robots.

Fig. 2: Execution-interface dynamics adaptation (EIDA). Target-system execution data identify independent pose-increment and velocity-feedback models. The execution-fitted interface supports policy training in simulation; at deployment, live sensing and the existing controller replace the fitted responses.

Fig. 3: Pose-prediction accuracy on the Jackal and Go2 validation sets. Curves show mean endpoint RMSE over the prediction horizon; error bars indicate one standard deviation across three pose-model seeds.

Panel of Figure 4 — Fig. 4: Training and evaluation settings: (a) Simulation training map, (b) Jackal in BARN, and (c) physical Go2.

Method	SR (%) $\uparrow$	CR (%) $\downarrow$	TR (%) $\downarrow$	Metric $\uparrow$	Global
DWA	46.00	37.00	17.00	0.2130	Yes
E-Band	57.70	9.70	32.60	0.2880	Yes
Nominal	71.22 $\pm$ 5.68	26.89	1.89	0.3547	No
Nominal	80.78 $\pm$ 1.35	18.44	0.78	0.3971	Yes
RWM-U	69.56 $\pm$ 4.40	25.56	4.89	0.3469	No
RWM-U	73.78 $\pm$ 5.67	25.56	0.67	0.3680	Yes

TABLE I: Simulation-based performance comparison of different models using the BARN dataset.

Fig. 5: Trajectories of the robot controlled by (a) Nominal, (b) RWM-U, and (c) EIDA in one of the 100 BARN testing environments. “S” represents the starting point, and “G” represents the target point. The corresponding videos are provided in the supplementary material.

Model	Hist.	Feedback	Global path: No		Global path: Yes
Model	Hist.	Feedback	SR (%)	Metric	SR (%)	Metric
Nominal	No	Command	71.67	0.3555	81.00	0.3977
Nominal	Yes	Command	70.67	0.3525	83.33	0.4082
Nominal	No	ARX	70.67	0.3505	83.33	0.4109
Nominal	Yes	ARX	72.33	0.3611	76.00	0.3726
Fitted	No	Command	73.33	0.3622	82.00	0.3954

TABLE II: Ablation of motion modeling, velocity history, and feedback source on BARN.

Panel of Figure 8 — Fig. 6: Physical Go2 navigation in two static scenes. Panels (a)–(d) compare the Nominal baseline with EIDA; a dynamic scenario in which a pedestrian suddenly obstructs the path is shown separately in Fig. 7 .

Fig. 7: Go2 navigation with EIDA in Scene 2 when a pedestrian suddenly obstructs the path. Overlaid robot and pedestrian poses illustrate the motion sequence toward the marked goal. This demonstration is separate from the static-scene trials in Table III .

Policy	Collision $\downarrow$	Goal $\uparrow$	Path (m) $\downarrow$	Time (s) $\downarrow$
Scene 1: Curved passage
Nominal	9/10	1/10	6.81	12.93
EIDA	0/10	10/10	6.32 $\boldsymbol{\pm}$ 0.18	10.31 $\boldsymbol{\pm}$ 0.75
Scene 2: Separated obstacles
Nominal	7/10	3/10	6.71 $\pm$ 0.37	12.04 $\pm$ 2.29
EIDA	0/10	10/10	5.91 $\boldsymbol{\pm}$ 0.30	8.98 $\boldsymbol{\pm}$ 1.04

TABLE III: Real-world Go2 navigation performance in two scenes.

Explore similar work

Online Sim-to-Real Adaptation via Closed-Loop System Modeling
Sep 24, 2026Yuhao Huang, Samuel A. Moore, Boyuan ChenSim-To-Real Gap Closed-Loop Feedback
Duke University
Abstract Sim2Real through Approximate Information States
Apr 16, 2026Yunfu Deng, Yuhao Li, Josiah P. HannaSim-To-Real Reinforcement Learning Sim-To-Real Gap
Department of Computer Sciences, University of Wisconsin–Madison, Madison, WI 53706 USA · Manning College of Information and Computer Sciences, University of Massachusetts Amherst, Amherst, MA 01003 USA
EmbodiRSI: Recursive Self-Improvement for Data-Efficient Robot Adaptation
Sep 30, 2026Haoran Lang, Haotao Lu, Shiyu Sang +6Scalable Robot Learning Robotic Manipulation Policies
InstAdapt · ShanghaiTech University

Sep 24, 2026, cs.ROJ/K move · Enter open · S save

Online Sim-to-Real Adaptation via Closed-Loop System Modeling

Yuhao Huang, Samuel A. Moore, Boyuan Chen

Duke University

Sim-To-Real Gap Closed-Loop Feedback Scalable Robot Learning

Open paper AskPDF arXiv

Sim-to-real transfer has made substantial progress, but can still produce controllers that remain stable and functional on hardware while suffering from degraded tracking accuracy due to residual dynamics mismatch. Correcting these errors typically requires identifying the underlying system dynamics, adapting the control policy, or returning to simulation for additional training and finetuning, all of which can require substantial data and computation. We propose OSRAM (Online Sim-to-Real Adaptation via Closed-Loop System Modeling), a framework that instead adapts the reference commands provided to an existing controller. OSRAM treats the deployed robot and its policy as a unified closed-loop dynamical system and learns its task-level command-response behavior directly from tracking observations. A closed-loop dynamics model is meta-trained across randomized dynamics in simulation and rapidly finetuned after deployment using limited real-world interaction. The adapted model is then used to optimize future reference commands while leaving the underlying control policy unchanged. We evaluate OSRAM on bipedal velocity tracking and loco-manipulation in simulation and on hardware. Results show that closed-loop modeling improves prediction and tracking accuracy under unseen dynamics, while online reference adaptation reduces residual sim-to-real tracking errors across different control objectives and hardware configurations. These results demonstrate that adapting the behavior of the robot-policy closed loop provides a practical alternative to finetuning the policy or identifying the full physical dynamics for sim-to-real transfer. More information can be found at http://generalroboticslab.com/OSRAM.

Abstract

Explore similar work

Online Sim-to-Real Adaptation via Closed-Loop System Modeling

Abstract Sim2Real through Approximate Information States

EmbodiRSI: Recursive Self-Improvement for Data-Efficient Robot Adaptation