---
title: "Reinforcement Learning for Robotic Manipulation: From Simulation to Real World"
date: 2023-12-15T00:00:00Z
draft: false
authors: "J. Smith, M. Chen, K. Patel"
journal: "Robotics and Autonomous Systems"
volume: "168"
pages: "104-121"
doi: "10.1016/j.robot.2023.104512"
pdf: "/papers/rl-robotics.pdf"
tags: ["reinforcement-learning", "robotics", "sim-to-real"]
---

We present a novel approach to bridging the simulation-to-reality gap in robotic manipulation tasks using domain randomization and adaptive policy transfer.

## Abstract

Transferring reinforcement learning policies from simulation to real-world robotic systems remains a significant challenge. This paper introduces SimAdapt, a framework that combines domain randomization with online adaptation to achieve robust policy transfer. We demonstrate our approach on a set of manipulation tasks and show significant improvements over baseline methods in terms of success rate and sample efficiency.