clvrai / mopa-rl Goto Github PK
View Code? Open in Web Editor NEWMotion Planner Augmented Reinforcement Learning for Robot Manipulation in Obstructed Environments (CoRL 2020)
Home Page: https://clvrai.com/mopa-rl
License: MIT License
Motion Planner Augmented Reinforcement Learning for Robot Manipulation in Obstructed Environments (CoRL 2020)
Home Page: https://clvrai.com/mopa-rl
License: MIT License
Hi,
I'm trying to run the script scripts/2d/mopa.sh and I get the following segmentation fault, and I'm not really sure where to start with addressing this. If anyone has any suggestions, that would be greatly appreciated.
Thanks!
[a53:232685] *** Process received signal *** [a53:232685] Signal: Segmentation fault (11) [a53:232685] Signal code: Address not mapped (1) [a53:232685] Failing at address: (nil) [a53:232685] [ 0] /lib/x86_64-linux-gnu/libpthread.so.0(+0x14420)[0x7fc8b0d30420] [a53:232685] [ 1] /home/tarunc/mopa-rl/motion_planners/planner.cpython-37m-x86_64-linux-gnu.so(_ZN6MjOmpl26MujocoStateValidityChecker21addGlueTransformationERKSt6vectorINSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEESaIS7_EE+0x59)[0x7fc8300bf2a9] [a53:232685] [ 2] /home/tarunc/mopa-rl/motion_planners/planner.cpython-37m-x86_64-linux-gnu.so(_ZN13MotionPlanner16KinematicPlanner12isValidStateESt6vectorIdSaIdEE+0x1e9)[0x7fc8300bd3a9] [a53:232685] [ 3] /home/tarunc/mopa-rl/motion_planners/planner.cpython-37m-x86_64-linux-gnu.so(+0x41937)[0x7fc8300d3937] [a53:232685] [ 4] /home/tarunc/mopa-rl/motion_planners/planner.cpython-37m-x86_64-linux-gnu.so(+0x41e32)[0x7fc8300d3e32] [a53:232685] [ 5] python(_PyMethodDescr_FastCallKeywords+0xdb)[0x5643905dfd4b] [a53:232685] [ 6] python(+0x17faae)[0x5643905e0aae] [a53:232685] [ 7] python(_PyEval_EvalFrameDefault+0x661)[0x564390624601] [a53:232685] [ 8] python(_PyFunction_FastCallKeywords+0x187)[0x5643905998d7] [a53:232685] [ 9] python(+0x17f9c5)[0x5643905e09c5] [a53:232685] [10] python(_PyEval_EvalFrameDefault+0x661)[0x564390624601] [a53:232685] [11] python(_PyFunction_FastCallKeywords+0x187)[0x5643905998d7] [a53:232685] [12] python(+0x17f9c5)[0x5643905e09c5] [a53:232685] [13] python(_PyEval_EvalFrameDefault+0x661)[0x564390624601] [a53:232685] [14] python(_PyFunction_FastCallKeywords+0x187)[0x5643905998d7] [a53:232685] [15] python(+0x17f9c5)[0x5643905e09c5] [a53:232685] [16] python(_PyEval_EvalFrameDefault+0x661)[0x564390624601] [a53:232685] [17] python(+0x1901a5)[0x5643905f11a5] [a53:232685] [18] python(_PyMethodDef_RawFastCallKeywords+0xe9)[0x5643905aa639] [a53:232685] [19] python(_PyEval_EvalFrameDefault+0x4428)[0x5643906283c8] [a53:232685] [20] python(_PyFunction_FastCallKeywords+0x187)[0x5643905998d7] [a53:232685] [21] python(+0x17f9c5)[0x5643905e09c5] [a53:232685] [22] python(_PyEval_EvalFrameDefault+0x661)[0x564390624601] [a53:232685] [23] python(_PyFunction_FastCallKeywords+0x187)[0x5643905998d7] [a53:232685] [24] python(_PyEval_EvalFrameDefault+0x3f5)[0x564390624395] [a53:232685] [25] python(_PyEval_EvalCodeWithName+0x255)[0x564390579e85] [a53:232685] [26] python(+0x1d7d8e)[0x564390638d8e] [a53:232685] [27] python(_PyMethodDef_RawFastCallKeywords+0xe9)[0x5643905aa639] [a53:232685] [28] python(_PyEval_EvalFrameDefault+0x4428)[0x5643906283c8] [a53:232685] [29] python(_PyEval_EvalCodeWithName+0x255)[0x564390579e85] [a53:232685] *** End of error message *** Segmentation fault (core dumped)
Hi! I'm trying to set up the motion planner wrapper. I've cloned ompl (from the official source) into the mopa-rl directory, and run the setup commands. I then cd'ed into motion-planners and tried running the command, but ended up with this issue:
cc1plus: warning: command line option ‘-Wstrict-prototypes’ is valid for C/ObjC but not for C++ ./src/mujoco_ompl_interface.cpp:2:10: fatal error: ompl/base/StateSpace.h: No such file or directory 2 | #include <ompl/base/StateSpace.h> | ^~~~~~~~~~~~~~~~~~~~~~~~ compilation terminated. error: command '/usr/bin/gcc' failed with exit code 1
I'm not entirely sure where this is coming from, so if anyone had any insight, that would be greatly appreciated. Thanks!
Hi, thank you for your generous sharing. I have installed all the libraries and when I execute code
sh ./scripts/2d/baseline.sh
, I encounter the following problem
MoPA-RL: error: argument --gpu expected one argument.
Do u know how to fix it? Looking forward to your reply, thanks!
Hi!
In your paper,
After the motion planning, the resulting path is smoothed using a shortcutting algorithm.
is mentioned to smooth the resulting path with the algorithm in
R. Geraerts and M. H. Overmars. Creating high-quality paths for motion planning. The International Journal of Robotics Research, 26(8):845–863, 2007.
But I didn't find this post-process in this library.
Did I miss something?
Thanks!
Hi! Thanks for your generous sharing!
When I read the XML files, I noticed there are a lot of duplications of the body, such as the indicator link, dummy link or joint, goal link or joint.
What are the differences between the above settings and the real link or joint? In other words, what's the role of the above settings in the RL simulation?
Looking forward to your reply!
when try to run sh ./scripts/2d/mopa.sh, there will be sh: 2: Syntax error: "(" unexpected, ERROR: Invalid activation key. After checking, I think there is something wrong with PyKinematicPlanner. Could anyone can fix it?
Hi! It seems that there is no #include <GL/gl.h>
and #include <GL/glext.h>
in the env/lib/python3.7/site-packages/mujoco_py/gl/eglshim.c.
My mujoco_py version is 2.0.2.5 and the first 9 lines in eglshim.c
are:
#define EGL_EGLEXT_PROTOTYPES
#include "egl.h"
#include "eglext.h"
#include <GL/glew.h>
#include "mujoco.h"
#include "mjrender.h"
#include "glshim.h"
Looking forward to your reply! Thx!
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.