Get the latest tech news

DrEureka: Language Model Guided SIM-to-Real Transfer


policies learned in simulation to the real world is a promising strategy for acquiring robot skills at scale. However, sim-to-real approaches typically rely on manual design and tuning of the task reward function as well as the simulation physics parameters, rendering the process slow and human-labor intensive.

However, sim-to-real approaches typically rely on manual design and tuning of the task reward function as well as the simulation physics parameters, rendering the process slow and human-labor intensive. Our LLM-guided sim-to-real approach requires only the physics simulation for the target task and automatically constructs suitable reward functions and domain randomization distributions to support real-world transfer. Then, we showcase that our approach is capable of solving novel robot tasks, such as quadruped balancing and walking atop a yoga ball, without iterative manual design.

Get the Android app

Or read this on Hacker News

Read more on:

Photo of transfer

transfer

Photo of language model

language model

Photo of sim

sim

Related news:

News photo

Upcoming life-sim inZOI ready to take on The Sims at its own game

News photo

T-Mobile, Verizon workers get texts offering $300 for SIM swaps

News photo

SIM swap crooks solicit T-Mobile US, Verizon staff via text to do their dirty work