Comments (6)
We also found another issue in the joint space planner that can cause it to work with low accuracy / success. We are investigating a fix.
from curobo.
We have joint space planner working better in the latest release. Let us know if you have issues.
from curobo.
Additionally, do you have any suggestions on how to debug when there is no plan generated? I know it could it be a resource limitation, MotionGenConfig or PlanConfig or just the collision environment. Is there a way of knowing what's causing this? Thank you
from curobo.
Can you share the start_js as well? And is it for the Franka Panda robot?
We have a logger that will create an animated usd of the solver iterations. Though we haven't tested with plan_single_js
Here is an example:
curobo/examples/usd_example.py
Line 168 in 774dcfd
from curobo.
- I have a UR5e arm running in simulation. This is my start_js
Planning to 'home' position JointState(position=tensor([-0.4660, -1.9553, -1.6230, -2.7051, -2.0357, 3.1413], device='cuda:0'), velocity=tensor([-0., 0., 0., 0., 0., 0.], device='cuda:0'), acceleration=tensor([-0., 0., 0., 0., 0., 0.], device='cuda:0'), joint_names=['shoulder_pan_joint', 'shoulder_lift_joint', 'elbow_joint', 'wrist_1_joint', 'wrist_2_joint', 'wrist_3_joint'], jerk=tensor([-0., 0., 0., 0., 0., 0.], device='cuda:0'), tensor_args=TensorDeviceType(device=device(type='cuda', index=0), dtype=torch.float32, collision_geometry_dtype=torch.float32, collision_gradient_dtype=torch.float32, collision_distance_dtype=torch.float32))
-
I'm trying to reliably navigate between these 2 joint states for a ur5e: joint_state_1 = [1.5127, -1.8392, 1.87056, -0.1426, 3.15, 3.1165], joint_state_2 = [-1.588, -1.633, -2.104, -2.324, 1.661, 3.107]. There are no obstacles in my test environment.
-
Additionally, I just tried the simple_stacking.py isaac example which has the usdhelper calling write_motion_gen_log api. I switched on the save_log flag to True to print and save debug information and i get this error.
[7.206s] Simulation App Startup Complete
warming up...
Traceback (most recent call last):
File "examples/isaac_sim/simple_stacking.py", line 435, in
my_world=my_world, my_task=my_task, constrain_grasp_approach=args.constrain_grasp_approach
File "examples/isaac_sim/simple_stacking.py", line 161, in init
self.motion_gen.warmup(parallel_finetune=True)
File "curobo/src/curobo/wrap/reacher/motion_gen.py", line 2202, in warmup
self.plan_single_js(start_state, goal_state, MotionGenPlanConfig(max_attempts=1))
File "curobo/src/curobo/wrap/reacher/motion_gen.py", line 2372, in plan_single_js
solve_state, start_state, goal_state, plan_config=plan_config
File "curobo/src/curobo/wrap/reacher/motion_gen.py", line 1842, in _plan_js_from_solve_state
result.debug_info["trajopt_result"] = traj_result
TypeError: 'NoneType' object does not support item assignment
2024-04-08 20:54:16 [14,746ms] [Warning] [carb.audio.context] 1 contexts were leaked
/home/shannon/.local/share/ov/pkg/isaac_sim-2022.2.1/python.sh: line 41: 593397 Segmentation fault (core dumped)$python_exe "$ @" $args
There was an error running python
from curobo.
An update on 2) regarding navigating to the two joint states. There was a self collision in my tool joint causing the failure to plan. I reduced the sphere size to a very low value and it works reliably. I think I'll need to add this to the self_collision_ignore
. Would be helpful to see a debug message when self collision happens.
from curobo.
Related Issues (20)
- "qx, qy, qz, qw" or "qw, qx, qy, qz" HOT 1
- RuntimeError: CUDA error: an illegal memory access was encountered Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions HOT 9
- timeout in MotionGenPlanConfig not strictly enforced HOT 3
- attach_objects_to_robot/detach_object_from_robot HOT 1
- Installation problems for building with nvblox HOT 2
- Failing a 'base case' motion generation planning problem HOT 9
- Error building docker - nvblox_torch installation HOT 3
- Smooth the trajectory
- Error planning to goalset with constraints -- `QuatToMatrix` assumes 1 batch dim HOT 1
- How to reduce the length of trajectory HOT 2
- Velocity scale applied twice to limit HOT 1
- Nvblox collision checker demo sometimes gives incorrect results HOT 1
- Error- Installing with python HOT 1
- Element 0 of tensors does not require grad and does not have a grad_fn HOT 3
- "Getting Started Using in Python" Examples Not Working HOT 2
- Strange behavior of `update_world` not working unless the mesh names change HOT 12
- `motion_gen.warmup` resulting in breaking cuda graph when using `link_poses` with `plan_batch` HOT 1
- Occasional bug with `g_dim` shape HOT 3
- UserWarning occurred when testing motion generation HOT 1
- Gradient of `PoseError` becomes INF HOT 4
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
π 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. πππ
-
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 β€οΈ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from curobo.