jimyoung6709 commited on
Commit
916b800
·
verified ·
1 Parent(s): c4d9d02

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +6 -15
README.md CHANGED
@@ -6,14 +6,13 @@ tags:
6
  - robotics
7
  - motion planning
8
  ---
9
-
10
  # Neural MP
11
 
12
- Neural MP is a machine learning-based motion planning system for robotic manipulation tasks. It combines neural networks trained on large-scale simulated data with lightweight optimization techniques to generate efficient, collision-free trajectories. Neural MP is designed to generalize across diverse environments and obstacle configurations, making it suitable for both simulated and real-world robotic applications. This repository contains the implementation, data generation tools, and evaluation scripts for Neural MP.
13
 
14
- All Neural MP checkpoints, as well as our [training codebase](https://github.com/mihdalal/neuralmotionplanner) are released under an MIT License.
15
 
16
- For full details, please read our paper(coming soon) and see [our project page](https://mihdalal.github.io/neuralmotionplanner/).
17
 
18
  ## Model Summary
19
  - **Developed by:** The Neural MP team consisting of researchers from Carnegie Mellon University.
@@ -26,7 +25,7 @@ For full details, please read our paper(coming soon) and see [our project page](
26
 
27
  ## Installation
28
 
29
- Please check [here](https://github.com/mihdalal/neural_mp?tab=readme-ov-file#installation-instructions) for detailed instructions
30
 
31
  ## Usage
32
 
@@ -34,17 +33,14 @@ Neural MP model takes in 3D point cloud and start & goal angles of the Franka ro
34
 
35
  Here's an deployment example with the Manimo Franka control library:
36
 
37
- Note: using Manimo is not required, you may use other Franka control libraries by creating a wrapper class which inherits from FrankaRealEnv (check [franka_real_env.py](https://github.com/mihdalal/neural_mp/blob/master/neural_mp/envs/franka_real_env.py))
38
 
39
  ```python
40
  import argparse
41
  import numpy as np
42
-
43
  from neural_mp.envs.franka_real_env import FrankaRealEnvManimo
44
  from neural_mp.real_utils.neural_motion_planner import NeuralMP
45
-
46
  if __name__ == "__main__":
47
-
48
  parser = argparse.ArgumentParser()
49
  parser.add_argument(
50
  "--mdl_url",
@@ -89,7 +85,6 @@ if __name__ == "__main__":
89
  default=[0.1, 0.1, 0.1, 0.0, 0.0, 0.1, 0.0, 0.0, 0.0, 1.0],
90
  help="Specify the bounding box of the in hand object. 10 params in total [size(xyz), pos(xyz), ori(xyzw)] 3+3+4.",
91
  )
92
-
93
  args = parser.parse_args()
94
  env = FrankaRealEnvManimo()
95
  neural_mp = NeuralMP(
@@ -100,18 +95,15 @@ if __name__ == "__main__":
100
  in_hand_params=args.in_hand_params,
101
  visualize=True,
102
  )
103
-
104
  points, colors = neural_mp.get_scene_pcd(
105
  use_cache=args.use_cache,
106
  cache_name=args.cache_name,
107
  debug_combined_pcd=args.debug_combined_pcd,
108
  denoise=args.denoise_pcd,
109
  )
110
-
111
  # specify start and goal configurations
112
  start_config = np.array([-0.538, 0.628, -0.061, -1.750, 0.126, 2.418, 1.610])
113
  goal_config = np.array([1.067, 0.847, -0.591, -1.627, 0.623, 2.295, 2.580])
114
-
115
  if args.tto:
116
  trajectory = neural_mp.motion_plan_with_tto(
117
  start_config=start_config,
@@ -126,6 +118,5 @@ if __name__ == "__main__":
126
  points=points,
127
  colors=colors,
128
  )
129
-
130
  success, joint_error = neural_mp.execute_motion_plan(trajectory, speed=0.2)
131
- ```
 
6
  - robotics
7
  - motion planning
8
  ---
 
9
  # Neural MP
10
 
11
+ Neural MP is a machine learning-based motion planning system for robotic manipulation tasks. It combines neural networks trained on large-scale simulated data with lightweight optimization techniques to generate efficient, collision-free trajectories. Neural MP is designed to generalize across diverse environments and obstacle configurations, making it suitable for both simulated and real-world robotic applications. This repository contains the model weights for Neural MP.
12
 
13
+ All Neural MP checkpoints, as well as our [codebase](https://github.com/mihdalal/neuralmotionplanner) are released under an MIT License.
14
 
15
+ For full details, please read our [paper](https://mihdalal.github.io/neuralmotionplanner/resources/paper.pdf) and see [our project page](https://mihdalal.github.io/neuralmotionplanner/).
16
 
17
  ## Model Summary
18
  - **Developed by:** The Neural MP team consisting of researchers from Carnegie Mellon University.
 
25
 
26
  ## Installation
27
 
28
+ Please read [here](https://github.com/mihdalal/neural_mp?tab=readme-ov-file#installation-instructions) for detailed instructions
29
 
30
  ## Usage
31
 
 
33
 
34
  Here's an deployment example with the Manimo Franka control library:
35
 
36
+ Note: using Manimo is not required, you may use other Franka control libraries by creating a wrapper class which inherits from FrankaRealEnv (see [franka_real_env.py](https://github.com/mihdalal/neural_mp/blob/master/neural_mp/envs/franka_real_env.py))
37
 
38
  ```python
39
  import argparse
40
  import numpy as np
 
41
  from neural_mp.envs.franka_real_env import FrankaRealEnvManimo
42
  from neural_mp.real_utils.neural_motion_planner import NeuralMP
 
43
  if __name__ == "__main__":
 
44
  parser = argparse.ArgumentParser()
45
  parser.add_argument(
46
  "--mdl_url",
 
85
  default=[0.1, 0.1, 0.1, 0.0, 0.0, 0.1, 0.0, 0.0, 0.0, 1.0],
86
  help="Specify the bounding box of the in hand object. 10 params in total [size(xyz), pos(xyz), ori(xyzw)] 3+3+4.",
87
  )
 
88
  args = parser.parse_args()
89
  env = FrankaRealEnvManimo()
90
  neural_mp = NeuralMP(
 
95
  in_hand_params=args.in_hand_params,
96
  visualize=True,
97
  )
 
98
  points, colors = neural_mp.get_scene_pcd(
99
  use_cache=args.use_cache,
100
  cache_name=args.cache_name,
101
  debug_combined_pcd=args.debug_combined_pcd,
102
  denoise=args.denoise_pcd,
103
  )
 
104
  # specify start and goal configurations
105
  start_config = np.array([-0.538, 0.628, -0.061, -1.750, 0.126, 2.418, 1.610])
106
  goal_config = np.array([1.067, 0.847, -0.591, -1.627, 0.623, 2.295, 2.580])
 
107
  if args.tto:
108
  trajectory = neural_mp.motion_plan_with_tto(
109
  start_config=start_config,
 
118
  points=points,
119
  colors=colors,
120
  )
 
121
  success, joint_error = neural_mp.execute_motion_plan(trajectory, speed=0.2)
122
+ ```