teddy-liao/walk-these-ways-go2

Deploy walk-these-ways project on Unitree Go2

walk-these-ways-go2's Issues

Send Request Error

When we are connected to the dog, as shown by the ping, and the dog is in sport mode, we get a send request error:

(pytorch_env) gabriel@gabriel-Legion-Pro-7-16IRX8H:/workspaces/walk-these-ways-go2/go2_gym_deploy/build$ ping 192.168.123.161
PING 192.168.123.161 (192.168.123.161) 56(84) bytes of data.
64 bytes from 192.168.123.161: icmp_seq=1 ttl=64 time=0.062 ms
64 bytes from 192.168.123.161: icmp_seq=2 ttl=64 time=0.071 ms
64 bytes from 192.168.123.161: icmp_seq=3 ttl=64 time=0.077 ms
^C
--- 192.168.123.161 ping statistics ---
3 packets transmitted, 3 received, 0% packet loss, time 2051ms
rtt min/avg/max/mdev = 0.062/0.070/0.077/0.006 ms
(pytorch_env) gabriel@gabriel-Legion-Pro-7-16IRX8H:/workspaces/walk-these-ways-go2/go2_gym_deploy/build$ sudo ./lcm_position_go2 enp114s0
[sudo] password for gabriel:
Communication level is set to LOW-level.
WARNING: Make sure the robot is hung up.
Caution: The scripts is about to shutdown Unitree sport_mode Service.
Press Enter to continue...

send request error. requestId:7900295264479
sportd_mode is already deactivated now
next step is setting up communication
Press Enter to continue...

SET NOMINAL POSE
Communicatino is set up successfully
LCM <<<------------>>> Unitree SDK2

---

Press L2+B if any unexpected error occurs

Sport mode is not already deactivated, so we are not sure what is happening.

Question about simulation environment

Dear authors,

Thanks for your excellent work. I'd like to ask are there any referred code or repo that can achieve the real-time mapping of real Go2 in simulation environment?

编译unitree_sdk的时候报错

[ 2%] Linking CXX executable h1_low_level_example
//usr/local/lib/libunitree_sdk2.a(os.cpp.o)：在函数‘unitree::common::OsHelper::GetTid()’中：
os.cpp:(.text+0x4d3)：对‘gettid’未定义的引用
//usr/local/lib/libunitree_sdk2.a(lock.cpp.o)：在函数‘unitree::common::AssertEqual(int, int, bool, char const*, char const*, int)’中：
lock.cpp:(.text._ZN7unitree6common11AssertEqualEiibPKcS2_i[_ZN7unitree6common11AssertEqualEiibPKcS2_i]+0x84)：对‘gettid’未定义的引用
lock.cpp:(.text._ZN7unitree6common11AssertEqualEiibPKcS2_i[_ZN7unitree6common11AssertEqualEiibPKcS2_i]+0x190)：对‘gettid’未定义的引用
//usr/local/lib/libunitree_sdk2.a(lock.cpp.o)：在函数‘unitree::common::AssertEqual(int, int, int, bool, char const*, char const*, int)’中：
lock.cpp:(.text._ZN7unitree6common11AssertEqualEiiibPKcS2_i[_ZN7unitree6common11AssertEqualEiiibPKcS2_i]+0x9c)：对‘gettid’未定义的引用
lock.cpp:(.text._ZN7unitree6common11AssertEqualEiiibPKcS2_i[_ZN7unitree6common11AssertEqualEiiibPKcS2_i]+0x1a3)：对‘gettid’未定义的引用
//usr/local/lib/libunitree_sdk2.a(lock.cpp.o):lock.cpp:(.text._ZN7unitree6common13AssertEqualExEiiibPKcS2_i[_ZN7unitree6common13AssertEqualExEiiibPKcS2_i]+0xaf): 跟着更多未定义的参考到 gettid
//usr/local/lib/libunitree_sdk2.a(recurrent_thread.cpp.o)：在函数‘unitree::common::Exception::ToStringabi:cxx11 const’中：
recurrent_thread.cpp:(.text._ZNK7unitree6common9Exception8ToStringB5cxx11Ev[_ZNK7unitree6common9Exception8ToStringB5cxx11Ev]+0x38)：对‘std::__cxx11::basic_ostringstream<char, std::char_traits, std::allocator >::basic_ostringstream()’未定义的引用
//usr/local/lib/libunitree_sdk2.a(recurrent_thread.cpp.o)：在函数‘unitree::common::AssertNotEqual(int, int, bool, char const*, char const*, int)’中：
recurrent_thread.cpp:(.text._ZN7unitree6common14AssertNotEqualEiibPKcS2_i[_ZN7unitree6common14AssertNotEqualEiibPKcS2_i]+0x84)：对‘gettid’未定义的引用
recurrent_thread.cpp:(.text._ZN7unitree6common14AssertNotEqualEiibPKcS2_i[_ZN7unitree6common14AssertNotEqualEiibPKcS2_i]+0x190)：对‘gettid’未定义的引用
//usr/local/lib/libunitree_sdk2.a(dds_factory_model.cpp.o)：在函数‘void unitree::common::Logger::Log<char const (&) [36], unsigned int&>(int, char const (&) [36], unsigned int&)’中：
dds_factory_model.cpp:(.text.ZN7unitree6common6Logger3LogIJRA36_KcRjEEEviDpOT[ZN7unitree6common6Logger3LogIJRA36_KcRjEEEviDpOT]+0x88)：对‘std::__cxx11::basic_ostringstream<char, std::char_traits, std::allocator >::basic_ostringstream()’未定义的引用
//usr/local/lib/libunitree_sdk2.a(dds_factory_model.cpp.o)：在函数‘void unitree::common::Logger::Log<char const (&) [36]>(int, char const (&) [36])’中：
dds_factory_model.cpp:(.text.ZN7unitree6common6Logger3LogIJRA36_KcEEEviDpOT[ZN7unitree6common6Logger3LogIJRA36_KcEEEviDpOT]+0x81)：对‘std::__cxx11::basic_ostringstream<char, std::char_traits, std::allocator >::basic_ostringstream()’未定义的引用
//usr/local/lib/libddscxx.so：对‘std::__cxx11::basic_stringstream<char, std::char_traits, std::allocator >::basic_stringstream()@GLIBCXX_3.4.26’未定义的引用
collect2: error: ld returned 1 exit status
CMakeFiles/h1_low_level_example.dir/build.make:94: recipe for target 'h1_low_level_example' failed
make[2]: *** [h1_low_level_example] Error 1
CMakeFiles/Makefile2:67: recipe for target 'CMakeFiles/h1_low_level_example.dir/all' failed
make[1]: *** [CMakeFiles/h1_low_level_example.dir/all] Error 2
Makefile:83: recipe for target 'all' failed
make: *** [all] Error 2

Questions about Sdk

Tysm for sharing this repo and the detailed comments! I’ve just started working with unitree robot and was wondering if you’d have some time to chat. I have some questions about the reason for LCM and the sdk itself and your help would be greatly appreciated. Lmk if I can contact you via email or wechat.

python play.py : Segmentation fault (core dumped)

(myconda) root@q80JeX:~/Downloads/walk-these-ways-go2-main/scripts# python play.py
Importing module 'gym_38' (/root/Downloads/isaacgym/python/isaacgym/_bindings/linux-x86_64/gym_38.so)
Setting GYM_USD_PLUG_INFO_PATH to /root/Downloads/isaacgym/python/isaacgym/_bindings/linux-x86_64/usd/plugInfo.json
PyTorch version 1.11.0+cu113
Device count 1
/root/Downloads/isaacgym/python/isaacgym/_bindings/src/gymtorch
Using /root/.cache/torch_extensions/py38_cu113 as PyTorch extensions root...
Emitting ninja build file /root/.cache/torch_extensions/py38_cu113/gymtorch/build.ninja...
Building extension module gymtorch...
Allowing ninja to set a default number of workers... (overridable by setting the environment variable MAX_JOBS=N)
ninja: no work to do.
Loading extension module gymtorch...
ML_LOGGER_USER is not set. This is required for online usage.
dict_keys(['AC_Args', 'PPO_Args', 'RunnerArgs', 'Cfg'])
dict_keys(['env', 'terrain', 'commands', 'curriculum_thresholds', 'init_state', 'control', 'asset', 'domain_rand', 'rewards', 'reward_scales', 'normalization', 'obs_scales', 'noise', 'noise_scales', 'viewer', 'sim', 'command_ranges'])
[Warning] [carb.gym.plugin] useGpuPipeline is set, forcing GPU PhysX
Not connected to PVD
+++ Using GPU PhysX
Physics Engine: PhysX
Physics Device: cuda:0
GPU Pipeline: enabled
Segmentation fault (core dumped)

请问运行时遇到了这个报错怎么解决？

About actuator network

Hello, thanks for your wonderful work!
Your work is helping me a lot, and I have a question.

Is the weight file of the actuator network (resources/actuator_nets/unitree_go1.pt) for unitree-go2? (Is the network trained using the data from unitree-go2?)

jetson and isaac gym

Is there any difference between the installation of isaac Gym in Jetson and in PC? My system will report an error
ImportError:/home/rl/IsaacGym/isaacGym/Python/isaacGym/unbinding/linux-x86_64/gym.38. so: cannot open shared object file: No such file or directory

A suggestion for git

Hi authors,

Thanks for your great work and make it available for the community.
I suggest that you can use .gitignore to ignore the update of some cache files like pycache. So that it will be more clear for us to identify your update.

Best