Source_Obj stringlengths 51 69 | Target_Obj stringlengths 51 69 | Thinking stringlengths 669 810 | Object stringlengths 140 285 | instruction stringlengths 41 76 | solution sequencelengths 6 6 | Conversation listlengths 2 2 |
|---|---|---|---|---|---|---|
[{"<|magenta|><|cube|>": ["<|12-18|><|local-0-1|>", 17]}] | [{"<|black|><|moon|>": ["<|6-11|><|local-3-0|>", 18]}] | LOCATE OBJECTS:
Target Object: <|magenta|><|cube|> found at <|12-18|><|local-0-1|> with height 17
Target Placement Location: <|black|><|moon|> at <|6-11|><|local-3-0|> with height 18
PLAN ACTIONS:
Step 1: Move above <|magenta|><|cube|> at <|12-18|><|local-0-1|> with height 30
Step 2: Lower to <|magenta|><|cube|> at <|12-18|><|local-0-1|> with height 17
Step 3: Close gripper on <|magenta|><|cube|> at <|12-18|><|local-0-1|> with height 17
Step 4: Lift <|magenta|><|cube|> at <|12-18|><|local-0-1|> to height 30
Step 5: Move to <|black|><|moon|> location at <|6-11|><|local-3-0|> with height 30
Step 6: Stack on top of <|black|><|moon|> location at <|6-11|><|local-3-0|> with height 18 + 5 and open gripper
| [{"violet-cube": [19, 5, 18]}, {"black-moon": [27, 44, 18]}, {"maroon-triangular prism": [9, 35, 17]}, {"magenta-cube": [48, 73, 17]}, {"rose-cube": [10, 92, 17]}, {"cyan-cube": [18, 49, 17]}, {"lime-cube": [64, 5, 18]}] | Stack the magenta cube on top of the black moon | [
[
48,
73,
30,
0,
60,
90,
1
],
[
48,
73,
17,
0,
60,
90,
1
],
[
48,
73,
17,
0,
60,
90,
0
],
[
48,
73,
30,
0,
60,
90,
0
],
[
27,
44,
30,
0,
60,
90,
0
],
[
... | [
{
"content": "You are a spatial reasoning assistant for a Franka Panda robot with a parallel gripper. Your task is to generate precise action sequences to accomplish object manipulation tasks.\n\n## INPUT ENVIRONMENT:\n- The workspace is a table surface represented as a 100x100 discrete grid, divided into a 25x... |
[{"<|magenta|><|cube|>": ["<|12-3|><|local-1-3|>", 17]}] | [{"<|rose|><|triangular prism|>": ["<|4-7|><|local-3-1|>", 18]}] | LOCATE OBJECTS:
Target Object: <|magenta|><|cube|> found at <|12-3|><|local-1-3|> with height 17
Target Placement Location: <|rose|><|triangular prism|> at <|4-7|><|local-3-1|> with height 18
PLAN ACTIONS:
Step 1: Move above <|magenta|><|cube|> at <|12-3|><|local-1-3|> with height 30
Step 2: Lower to <|magenta|><|cube|> at <|12-3|><|local-1-3|> with height 17
Step 3: Close gripper on <|magenta|><|cube|> at <|12-3|><|local-1-3|> with height 17
Step 4: Lift <|magenta|><|cube|> at <|12-3|><|local-1-3|> to height 30
Step 5: Move to <|rose|><|triangular prism|> location at <|4-7|><|local-3-1|> with height 30
Step 6: Stack on top of <|rose|><|triangular prism|> location at <|4-7|><|local-3-1|> with height 18 + 5 and open gripper
| [{"violet-star": [51, 38, 17]}, {"rose-star": [63, 35, 18]}, {"blue-triangular prism": [84, 33, 18]}, {"black-moon": [4, 31, 18]}, {"rose-moon": [8, 91, 18]}, {"rose-triangular prism": [19, 29, 18]}, {"magenta-cube": [49, 15, 17]}] | Stack the magenta cube on top of the rose triangular prism | [
[
49,
15,
30,
0,
60,
90,
1
],
[
49,
15,
17,
0,
60,
90,
1
],
[
49,
15,
17,
0,
60,
90,
0
],
[
49,
15,
30,
0,
60,
90,
0
],
[
19,
29,
30,
0,
60,
90,
0
],
[
... | [
{
"content": "You are a spatial reasoning assistant for a Franka Panda robot with a parallel gripper. Your task is to generate precise action sequences to accomplish object manipulation tasks.\n\n## INPUT ENVIRONMENT:\n- The workspace is a table surface represented as a 100x100 discrete grid, divided into a 25x... |
[{"<|black|><|cube|>": ["<|21-14|><|local-3-3|>", 19]}] | [{"<|rose|><|container|>": ["<|15-22|><|local-0-2|>", 19]}] | LOCATE OBJECTS:
Target Object: <|black|><|cube|> found at <|21-14|><|local-3-3|> with height 19
Target Placement Location: <|rose|><|container|> at <|15-22|><|local-0-2|> with height 19
PLAN ACTIONS:
Step 1: Move above <|black|><|cube|> at <|21-14|><|local-3-3|> with height 30
Step 2: Lower to <|black|><|cube|> at <|21-14|><|local-3-3|> with height 19
Step 3: Close gripper on <|black|><|cube|> at <|21-14|><|local-3-3|> with height 19
Step 4: Lift <|black|><|cube|> at <|21-14|><|local-3-3|> to height 30
Step 5: Move to <|rose|><|container|> location at <|15-22|><|local-0-2|> with height 30
Step 6: Place inside <|rose|><|container|> location at <|15-22|><|local-0-2|> with height 19 and open gripper
| [{"rose-container": [60, 90, 19]}, {"olive-cube": [4, 75, 17]}, {"silver-moon": [94, 19, 17]}, {"black-cube": [87, 59, 19]}, {"white-triangular prism": [77, 18, 17]}, {"gray-cylinder": [54, 63, 17]}] | Pick up the black cube and place it into the rose container | [
[
87,
59,
30,
0,
60,
90,
1
],
[
87,
59,
19,
0,
60,
90,
1
],
[
87,
59,
19,
0,
60,
90,
0
],
[
87,
59,
30,
0,
60,
90,
0
],
[
60,
90,
30,
0,
60,
90,
0
],
[
... | [
{
"content": "You are a spatial reasoning assistant for a Franka Panda robot with a parallel gripper. Your task is to generate precise action sequences to accomplish object manipulation tasks.\n\n## INPUT ENVIRONMENT:\n- The workspace is a table surface represented as a 100x100 discrete grid, divided into a 25x... |
[{"<|magenta|><|cylinder|>": ["<|19-11|><|local-3-0|>", 17]}] | [{"<|yellow|><|triangular prism|>": ["<|8-3|><|local-2-0|>", 18]}] | LOCATE OBJECTS:
Target Object: <|magenta|><|cylinder|> found at <|19-11|><|local-3-0|> with height 17
Target Placement Location: <|yellow|><|triangular prism|> at <|8-3|><|local-2-0|> with height 18
PLAN ACTIONS:
Step 1: Move above <|magenta|><|cylinder|> at <|19-11|><|local-3-0|> with height 30
Step 2: Lower to <|magenta|><|cylinder|> at <|19-11|><|local-3-0|> with height 17
Step 3: Close gripper on <|magenta|><|cylinder|> at <|19-11|><|local-3-0|> with height 17
Step 4: Lift <|magenta|><|cylinder|> at <|19-11|><|local-3-0|> to height 30
Step 5: Move to <|yellow|><|triangular prism|> location at <|8-3|><|local-2-0|> with height 30
Step 6: Stack on top of <|yellow|><|triangular prism|> location at <|8-3|><|local-2-0|> with height 18 + 5 and open gripper
| [{"azure-cylinder": [37, 65, 17]}, {"cyan-triangular prism": [12, 72, 17]}, {"magenta-cylinder": [79, 44, 17]}, {"cyan-cube": [60, 62, 18]}, {"yellow-triangular prism": [34, 12, 18]}, {"black-triangular prism": [52, 69, 17]}, {"purple-cylinder": [9, 20, 18]}] | Stack the magenta cylinder on top of the yellow triangular prism | [
[
79,
44,
30,
0,
60,
90,
1
],
[
79,
44,
17,
0,
60,
90,
1
],
[
79,
44,
17,
0,
60,
90,
0
],
[
79,
44,
30,
0,
60,
90,
0
],
[
34,
12,
30,
0,
60,
90,
0
],
[
... | [
{
"content": "You are a spatial reasoning assistant for a Franka Panda robot with a parallel gripper. Your task is to generate precise action sequences to accomplish object manipulation tasks.\n\n## INPUT ENVIRONMENT:\n- The workspace is a table surface represented as a 100x100 discrete grid, divided into a 25x... |
[{"<|gray|><|cylinder|>": ["<|6-20|><|local-3-3|>", 19]}] | [{"<|rose|><|cube|>": ["<|8-20|><|local-3-3|>", 18]}] | LOCATE OBJECTS:
Target Object: <|gray|><|cylinder|> found at <|6-20|><|local-3-3|> with height 19
Target Placement Location: <|rose|><|cube|> at <|8-20|><|local-3-3|> with height 18
PLAN ACTIONS:
Step 1: Move above <|gray|><|cylinder|> at <|6-20|><|local-3-3|> with height 30
Step 2: Lower to <|gray|><|cylinder|> at <|6-20|><|local-3-3|> with height 19
Step 3: Close gripper on <|gray|><|cylinder|> at <|6-20|><|local-3-3|> with height 19
Step 4: Lift <|gray|><|cylinder|> at <|6-20|><|local-3-3|> to height 30
Step 5: Move to <|rose|><|cube|> location at <|8-20|><|local-3-3|> with height 30
Step 6: Stack on top of <|rose|><|cube|> location at <|8-20|><|local-3-3|> with height 18 + 5 and open gripper
| [{"lime-cylinder": [19, 17, 18]}, {"navy-cube": [43, 30, 18]}, {"gray-cylinder": [27, 83, 19]}, {"silver-triangular prism": [34, 67, 18]}, {"rose-cube": [35, 83, 18]}] | Stack the gray cylinder on top of the rose cube | [
[
27,
83,
30,
0,
60,
90,
1
],
[
27,
83,
19,
0,
60,
90,
1
],
[
27,
83,
19,
0,
60,
90,
0
],
[
27,
83,
30,
0,
60,
90,
0
],
[
35,
83,
30,
0,
60,
90,
0
],
[
... | [
{
"content": "You are a spatial reasoning assistant for a Franka Panda robot with a parallel gripper. Your task is to generate precise action sequences to accomplish object manipulation tasks.\n\n## INPUT ENVIRONMENT:\n- The workspace is a table surface represented as a 100x100 discrete grid, divided into a 25x... |
[{"<|rose|><|cube|>": ["<|3-2|><|local-2-3|>", 18]}] | [{"<|red|><|moon|>": ["<|4-21|><|local-0-2|>", 17]}] | LOCATE OBJECTS:
Target Object: <|rose|><|cube|> found at <|3-2|><|local-2-3|> with height 18
Target Placement Location: <|red|><|moon|> at <|4-21|><|local-0-2|> with height 17
PLAN ACTIONS:
Step 1: Move above <|rose|><|cube|> at <|3-2|><|local-2-3|> with height 30
Step 2: Lower to <|rose|><|cube|> at <|3-2|><|local-2-3|> with height 18
Step 3: Close gripper on <|rose|><|cube|> at <|3-2|><|local-2-3|> with height 18
Step 4: Lift <|rose|><|cube|> at <|3-2|><|local-2-3|> to height 30
Step 5: Move to <|red|><|moon|> location at <|4-21|><|local-0-2|> with height 30
Step 6: Stack on top of <|red|><|moon|> location at <|4-21|><|local-0-2|> with height 17 + 5 and open gripper
| [{"red-moon": [16, 86, 17]}, {"rose-cube": [14, 11, 18]}, {"lime-cube": [25, 25, 18]}, {"purple-moon": [60, 41, 17]}, {"white-moon": [67, 62, 17]}] | Stack the rose cube on top of the red moon | [
[
14,
11,
30,
0,
60,
90,
1
],
[
14,
11,
18,
0,
60,
90,
1
],
[
14,
11,
18,
0,
60,
90,
0
],
[
14,
11,
30,
0,
60,
90,
0
],
[
16,
86,
30,
0,
60,
90,
0
],
[
... | [
{
"content": "You are a spatial reasoning assistant for a Franka Panda robot with a parallel gripper. Your task is to generate precise action sequences to accomplish object manipulation tasks.\n\n## INPUT ENVIRONMENT:\n- The workspace is a table surface represented as a 100x100 discrete grid, divided into a 25x... |
[{"<|purple|><|cylinder|>": ["<|15-5|><|local-3-3|>", 19]}] | [{"<|white|><|star|>": ["<|9-6|><|local-2-1|>", 19]}] | LOCATE OBJECTS:
Target Object: <|purple|><|cylinder|> found at <|15-5|><|local-3-3|> with height 19
Target Placement Location: <|white|><|star|> at <|9-6|><|local-2-1|> with height 19
PLAN ACTIONS:
Step 1: Move above <|purple|><|cylinder|> at <|15-5|><|local-3-3|> with height 30
Step 2: Lower to <|purple|><|cylinder|> at <|15-5|><|local-3-3|> with height 19
Step 3: Close gripper on <|purple|><|cylinder|> at <|15-5|><|local-3-3|> with height 19
Step 4: Lift <|purple|><|cylinder|> at <|15-5|><|local-3-3|> to height 30
Step 5: Move to <|white|><|star|> location at <|9-6|><|local-2-1|> with height 30
Step 6: Stack on top of <|white|><|star|> location at <|9-6|><|local-2-1|> with height 19 + 5 and open gripper
| [{"cyan-triangular prism": [95, 93, 18]}, {"white-star": [38, 25, 19]}, {"purple-cylinder": [63, 23, 19]}, {"black-cube": [28, 60, 17]}, {"purple-cube": [51, 1, 18]}, {"white-cube": [32, 44, 17]}, {"lime-star": [63, 29, 17]}] | Stack the purple cylinder on top of the white star | [
[
63,
23,
30,
0,
60,
90,
1
],
[
63,
23,
19,
0,
60,
90,
1
],
[
63,
23,
19,
0,
60,
90,
0
],
[
63,
23,
30,
0,
60,
90,
0
],
[
38,
25,
30,
0,
60,
90,
0
],
[
... | [
{
"content": "You are a spatial reasoning assistant for a Franka Panda robot with a parallel gripper. Your task is to generate precise action sequences to accomplish object manipulation tasks.\n\n## INPUT ENVIRONMENT:\n- The workspace is a table surface represented as a 100x100 discrete grid, divided into a 25x... |
[{"<|rose|><|moon|>": ["<|11-22|><|local-1-3|>", 18]}] | [{"<|cyan|><|container|>": ["<|8-11|><|local-3-1|>", 18]}] | LOCATE OBJECTS:
Target Object: <|rose|><|moon|> found at <|11-22|><|local-1-3|> with height 18
Target Placement Location: <|cyan|><|container|> at <|8-11|><|local-3-1|> with height 18
PLAN ACTIONS:
Step 1: Move above <|rose|><|moon|> at <|11-22|><|local-1-3|> with height 30
Step 2: Lower to <|rose|><|moon|> at <|11-22|><|local-1-3|> with height 18
Step 3: Close gripper on <|rose|><|moon|> at <|11-22|><|local-1-3|> with height 18
Step 4: Lift <|rose|><|moon|> at <|11-22|><|local-1-3|> to height 30
Step 5: Move to <|cyan|><|container|> location at <|8-11|><|local-3-1|> with height 30
Step 6: Place inside <|cyan|><|container|> location at <|8-11|><|local-3-1|> with height 18 and open gripper
| [{"navy-star": [16, 95, 18]}, {"magenta-cube": [3, 87, 17]}, {"rose-moon": [45, 91, 18]}, {"cyan-container": [35, 45, 18]}, {"purple-triangular prism": [59, 14, 17]}, {"red-cylinder": [3, 33, 18]}] | Pick up the rose moon and place it into the cyan container | [
[
45,
91,
30,
0,
60,
90,
1
],
[
45,
91,
18,
0,
60,
90,
1
],
[
45,
91,
18,
0,
60,
90,
0
],
[
45,
91,
30,
0,
60,
90,
0
],
[
35,
45,
30,
0,
60,
90,
0
],
[
... | [
{
"content": "You are a spatial reasoning assistant for a Franka Panda robot with a parallel gripper. Your task is to generate precise action sequences to accomplish object manipulation tasks.\n\n## INPUT ENVIRONMENT:\n- The workspace is a table surface represented as a 100x100 discrete grid, divided into a 25x... |
[{"<|black|><|moon|>": ["<|3-9|><|local-3-1|>", 19]}] | [{"<|magenta|><|container|>": ["<|5-6|><|local-3-0|>", 17]}] | LOCATE OBJECTS:
Target Object: <|black|><|moon|> found at <|3-9|><|local-3-1|> with height 19
Target Placement Location: <|magenta|><|container|> at <|5-6|><|local-3-0|> with height 17
PLAN ACTIONS:
Step 1: Move above <|black|><|moon|> at <|3-9|><|local-3-1|> with height 30
Step 2: Lower to <|black|><|moon|> at <|3-9|><|local-3-1|> with height 19
Step 3: Close gripper on <|black|><|moon|> at <|3-9|><|local-3-1|> with height 19
Step 4: Lift <|black|><|moon|> at <|3-9|><|local-3-1|> to height 30
Step 5: Move to <|magenta|><|container|> location at <|5-6|><|local-3-0|> with height 30
Step 6: Place inside <|magenta|><|container|> location at <|5-6|><|local-3-0|> with height 17 and open gripper
| [{"gray-star": [25, 64, 18]}, {"magenta-container": [23, 24, 17]}, {"olive-cube": [31, 49, 18]}, {"green-moon": [53, 89, 17]}, {"black-moon": [15, 37, 19]}, {"navy-triangular prism": [16, 49, 18]}, {"rose-star": [9, 34, 17]}] | Pick up the black moon and place it into the magenta container | [
[
15,
37,
30,
0,
60,
90,
1
],
[
15,
37,
19,
0,
60,
90,
1
],
[
15,
37,
19,
0,
60,
90,
0
],
[
15,
37,
30,
0,
60,
90,
0
],
[
23,
24,
30,
0,
60,
90,
0
],
[
... | [
{
"content": "You are a spatial reasoning assistant for a Franka Panda robot with a parallel gripper. Your task is to generate precise action sequences to accomplish object manipulation tasks.\n\n## INPUT ENVIRONMENT:\n- The workspace is a table surface represented as a 100x100 discrete grid, divided into a 25x... |
[{"<|red|><|cylinder|>": ["<|5-19|><|local-1-0|>", 17]}] | [{"<|gray|><|cylinder|>": ["<|17-18|><|local-2-1|>", 17]}] | LOCATE OBJECTS:
Target Object: <|red|><|cylinder|> found at <|5-19|><|local-1-0|> with height 17
Target Placement Location: <|gray|><|cylinder|> at <|17-18|><|local-2-1|> with height 17
PLAN ACTIONS:
Step 1: Move above <|red|><|cylinder|> at <|5-19|><|local-1-0|> with height 30
Step 2: Lower to <|red|><|cylinder|> at <|5-19|><|local-1-0|> with height 17
Step 3: Close gripper on <|red|><|cylinder|> at <|5-19|><|local-1-0|> with height 17
Step 4: Lift <|red|><|cylinder|> at <|5-19|><|local-1-0|> to height 30
Step 5: Move to <|gray|><|cylinder|> location at <|17-18|><|local-2-1|> with height 30
Step 6: Stack on top of <|gray|><|cylinder|> location at <|17-18|><|local-2-1|> with height 17 + 5 and open gripper
| [{"blue-cylinder": [42, 13, 17]}, {"rose-cylinder": [56, 66, 17]}, {"azure-cylinder": [74, 49, 17]}, {"purple-moon": [19, 28, 17]}, {"red-cylinder": [21, 76, 17]}, {"gray-cylinder": [70, 73, 17]}, {"green-star": [40, 40, 18]}] | Stack the red cylinder on top of the gray cylinder | [
[
21,
76,
30,
0,
60,
90,
1
],
[
21,
76,
17,
0,
60,
90,
1
],
[
21,
76,
17,
0,
60,
90,
0
],
[
21,
76,
30,
0,
60,
90,
0
],
[
70,
73,
30,
0,
60,
90,
0
],
[
... | [
{
"content": "You are a spatial reasoning assistant for a Franka Panda robot with a parallel gripper. Your task is to generate precise action sequences to accomplish object manipulation tasks.\n\n## INPUT ENVIRONMENT:\n- The workspace is a table surface represented as a 100x100 discrete grid, divided into a 25x... |
[{"<|violet|><|cube|>": ["<|22-16|><|local-2-3|>", 19]}] | [{"<|teal|><|container|>": ["<|8-19|><|local-3-2|>", 19]}] | LOCATE OBJECTS:
Target Object: <|violet|><|cube|> found at <|22-16|><|local-2-3|> with height 19
Target Placement Location: <|teal|><|container|> at <|8-19|><|local-3-2|> with height 19
PLAN ACTIONS:
Step 1: Move above <|violet|><|cube|> at <|22-16|><|local-2-3|> with height 30
Step 2: Lower to <|violet|><|cube|> at <|22-16|><|local-2-3|> with height 19
Step 3: Close gripper on <|violet|><|cube|> at <|22-16|><|local-2-3|> with height 19
Step 4: Lift <|violet|><|cube|> at <|22-16|><|local-2-3|> to height 30
Step 5: Move to <|teal|><|container|> location at <|8-19|><|local-3-2|> with height 30
Step 6: Place inside <|teal|><|container|> location at <|8-19|><|local-3-2|> with height 19 and open gripper
| [{"cyan-star": [10, 77, 18]}, {"violet-cube": [90, 67, 19]}, {"rose-cube": [34, 91, 17]}, {"teal-container": [35, 78, 19]}, {"cyan-moon": [1, 36, 18]}] | Pick up the violet cube and place it into the teal container | [
[
90,
67,
30,
0,
60,
90,
1
],
[
90,
67,
19,
0,
60,
90,
1
],
[
90,
67,
19,
0,
60,
90,
0
],
[
90,
67,
30,
0,
60,
90,
0
],
[
35,
78,
30,
0,
60,
90,
0
],
[
... | [
{
"content": "You are a spatial reasoning assistant for a Franka Panda robot with a parallel gripper. Your task is to generate precise action sequences to accomplish object manipulation tasks.\n\n## INPUT ENVIRONMENT:\n- The workspace is a table surface represented as a 100x100 discrete grid, divided into a 25x... |
[{"<|violet|><|cube|>": ["<|14-10|><|local-3-2|>", 18]}] | [{"<|cyan|><|triangular prism|>": ["<|0-1|><|local-0-3|>", 17]}] | LOCATE OBJECTS:
Target Object: <|violet|><|cube|> found at <|14-10|><|local-3-2|> with height 18
Target Placement Location: <|cyan|><|triangular prism|> at <|0-1|><|local-0-3|> with height 17
PLAN ACTIONS:
Step 1: Move above <|violet|><|cube|> at <|14-10|><|local-3-2|> with height 30
Step 2: Lower to <|violet|><|cube|> at <|14-10|><|local-3-2|> with height 18
Step 3: Close gripper on <|violet|><|cube|> at <|14-10|><|local-3-2|> with height 18
Step 4: Lift <|violet|><|cube|> at <|14-10|><|local-3-2|> to height 30
Step 5: Move to <|cyan|><|triangular prism|> location at <|0-1|><|local-0-3|> with height 30
Step 6: Stack on top of <|cyan|><|triangular prism|> location at <|0-1|><|local-0-3|> with height 17 + 5 and open gripper
| [{"cyan-triangular prism": [0, 7, 17]}, {"green-cube": [36, 68, 18]}, {"silver-cube": [80, 1, 18]}, {"violet-cube": [59, 42, 18]}, {"magenta-star": [96, 82, 17]}, {"gray-triangular prism": [2, 46, 18]}, {"teal-moon": [5, 52, 17]}] | Stack the violet cube on top of the cyan triangular prism | [
[
59,
42,
30,
0,
60,
90,
1
],
[
59,
42,
18,
0,
60,
90,
1
],
[
59,
42,
18,
0,
60,
90,
0
],
[
59,
42,
30,
0,
60,
90,
0
],
[
0,
7,
30,
0,
60,
90,
0
],
[
0,... | [
{
"content": "You are a spatial reasoning assistant for a Franka Panda robot with a parallel gripper. Your task is to generate precise action sequences to accomplish object manipulation tasks.\n\n## INPUT ENVIRONMENT:\n- The workspace is a table surface represented as a 100x100 discrete grid, divided into a 25x... |
End of preview. Expand in Data Studio
README.md exists but content is empty.
- Downloads last month
- 48