Comments (4)
Hello @abdul-mannan-khan, thank you for your interest in deepbots!
Since deepbots targets mainly the Webots simulation we cannot provide specific instructions on how you can transfer it to real robots, etc. Deepbots is intrinsically tied to Webots because this is its purpose. Depending on your use-case, you might find that setups with emitter-receiver can be easier to translate to physical systems as you are already doing. It heavily depends on your use-case.
However, assuming you want to run inference on a real robot using some generic system and a simulation-trained agent, you can strip down the environment you have created using deepbots, and replace the methods that for example grab the sensor data and run the motors (observation, action), load the trained agent and run inference using the observation data so you have a simple form of the RL loop deployed. Again, the specifics are heavily depend on your use-case, e.g. it will be quite different to do this using ROS to using a Lego Mindstorms robot.
I hope this proves helpful!
from deepbots.
@abdul-mannan-khan Hi, thanks for opening the issue. 😄
We have some tutorials here.
https://github.com/aidudezzz/deepbots-tutorials
Feel free to let us know if you need any assistance. We'll be happy to help.
from deepbots.
Hello @KelvinYang0320, Thank you for your response. I have gone through the tutorials (https://github.com/aidudezzz/deepbots-tutorials). However, I am still confused about the Sim2Real transformation procedure. Do you have any examples in it that describe how to shift Deepbots based RL algorithm to a physical system? In general, we connect with the simulation engine through some connection by calling a function (like Gym PyBulle engine). Our actions are executed by the simulator engine, and then when our training is finished, instead of calling the simulation engine, we connect it to the physical system. However, based on my understanding, we are just making a webot (.wbt) file where we are setting all the configuration, which calls WeBots. How can I shift it from calling WeBots to a physical system?
Thank you for your help. I am very grateful to you.
from deepbots.
@abdul-mannan-khan Umm, I don't have experience with transfering a deepbots to a physical system.
@eakirtas @tsampazk Could you share some ideas? Thank you!
from deepbots.
Related Issues (20)
- Create documentation site
- Modify Docker
- get_timestep deprecation warning in supervisor_emitter_receiver
- Is it possible to run several simulations in parallel? HOT 1
- RobotEmitterReceiver class should inherit from Webots Robot class
- doubt about the implement of the emitter-receiver scheme HOT 9
- Usage of snake_case
- A bug of robot never moving HOT 1
- CSV handle_emitter data types
- Error in gym setup command: 'extras_require' HOT 5
- Migrating to gymnasium
- Unable to install deepbots HOT 2
- Gym Environment and using stable baselines HOT 2
- There seems to be a bug in the step function of the robot_supervisor.py HOT 1
- Extend deepbots to support stable-baselines and implement gym-style default enviroments HOT 3
- Extend deepbots to support Evolutionary Algorithms HOT 14
- Question: how to get kinect camera information HOT 7
- [Tracker] Extension to Evolutionary Algorithms
- How to use ray to train in deepbots environment. HOT 3
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from deepbots.