# Analysis of Randomization Effects on Sim2Real Transfer in Reinforcement Learning for Robotic Manipulation Tasks

Source: [Analysis of Randomization Effects on Sim2Real Transfer in Reinforcement Learning for Robotic Manipulation Tasks](<https://www.sota2.com/research/paper/68fa6da21e209131cea3b089>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-04-30T05:00:04.703Z

Authors: Josip Josifovski, Mohammadhossein Malmir, Noah Klarmann, Bare Luka \\v{Z}agar, Nicol\\'as Navarro-Guerrero, Alois Knoll

Published: 2022-06-13T16:12:56.000Z

## Abstract

Randomization is currently a widely used approach in Sim2Real transfer for data-driven learning algorithms in robotics. Still, most Sim2Real studies report results for a specific randomization technique and often on a highly customized robotic system, making it difficult to evaluate different randomization approaches systematically. To address this problem, we define an easy-to-reproduce experimental setup for a robotic reach-and-balance manipulator task, which can serve as a benchmark for comparison. We compare four randomization strategies with three randomized parameters both in simulation and on a real robot. Our results show that more randomization helps in Sim2Real transfer, yet it can also harm the ability of the algorithm to find a good policy in simulation. Fully randomized simulations and fine-tuning show differentiated results and translate better to the real robot than the other approaches tested.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Grasping on Robot Grasping Simulation](<https://www.sota2.com/research/sota/grasping-on-robot-grasping-simulation.md>): Grasping; Robot Grasping Simulation; Episodic Reward: -12.43.

- [Robot Grasping Real World](<https://www.sota2.com/research/sota/robot-grasping-real-world.md>): Grasping; Robot Grasping Real World; Episodic Reward (rep): -14.99.

- [Reaching on Reacher Reality](<https://www.sota2.com/research/sota/reaching-on-reacher-reality.md>): Reaching; Reacher Reality; Completion Rate Score: -146.52.

- [Reaching on Reacher Simulation](<https://www.sota2.com/research/sota/reaching-on-reacher-simulation.md>): Reaching; Reacher Simulation; Repetition Metric: -129.16.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2206.06282>)

[PDF](<https://arxiv.org/pdf/2206.06282.pdf>)