Process for finding a path in an automated handling system, and handling system with corresponding control module for finding a path

Patent metadata

Field	Value
Publication number	US-9278452-B2
Application number	US-201514638668-A
Country	US
Kind code	B2
Filing date	Mar 4, 2015
Priority date	Mar 10, 2014
Publication date	Mar 8, 2016
Grant date	Mar 8, 2016

Abstract

An Automated handling system has a working area, at least one object in the working area, at least a first numerically controlled arm and at least a second numerically controlled arm. The arms are movable above the working area and have a collision avoidance mechanism. The current actual topography of the working area is determinable, a movement planning algorithm determines a path from a first position to a second position, wherein in the first application of the movement planning algorithm a reference path, a quasi-static collision examination for the path, which has been provided by the movement planning algorithm, is applicable, a dynamic collision examination for the path, which has been suggested by the movement planning algorithm, is applicable, and wherein the collision examination determine whether, when moving the first arm a collision would be provoked.

First claim

The invention claimed is:

1. Process ( 200 ) for collision avoidance in an automated handling system, which comprises:
a working area ( 10 ) which is configured for quasi-static receiving of at least one object (O 1 ),
at least two numerically controlled arms ( 11 , 12 ) which are movable above the working area ( 10 ),

wherein the following steps are carried out:
a. Providing the actual topography of the working area ( 10 ) including the current position of the at least one object (O 1 ),
b. Providing (S 3 . 1 ; S 2 , S 3 , S 4 . 1 ) a path from a first position (A) to a second position (B), wherein when this step b is passed through for the first time, a reference path, preferably the shortest path (W 1 ) is provided as a path,
c. Executing a quasi-static collision examination (S 4 , S 5 ; S 4 . 2 ) for the provided path (W 1 ), for determining whether the first of the two arms ( 11 ) would collide with the actual topography of the working area ( 10 ) when being moved along this path (W 1 ),
d. Repeating the steps b., c. and d. as long as in step c. a collision is detected, otherwise executing the subsequent step e.,
e. Executing a dynamic collision examination (S 4 , S 5 ; S 4 . 3 , S 5 . 1 ) for the path (W 1 ), for determining whether the arms ( 11 , 12 ) would collide with each other in case the first arm ( 11 ) would move along this path (W 1 );
f. Applying an evasion strategy if a collision was determined in step e., and executing a movement of the first arm ( 11 ) from the first position (A) to a second position (B) by applying the evasion strategy;
g. Executing a movement of the first arm ( 11 ) from the first position (A) to the second position (B) along the path (W 1 ) if the dynamic collision examination in step e. has not determined a collision.

2. Process according to claim 1 , wherein with each run through the step b. in each case another possible path from the first position (A) to the second position (B) is determined (S 2 , S 3 ; S 4 ) and provided, wherein the determination of further possible paths is based on a set of rules.

3. Process according to claim 2 , wherein in step b. other possible paths which leads from the first position (A) to the second position (B) are taken from a table or list ( 219 ) (S 3 . 1 ).

4. Process according to claim 3 , wherein in a preparatory step (S 2 . 2 ) multiple possible paths are determined by use of a set of rules and stored in the table or list ( 210 ).

5. Process according to claim 1 , wherein
i) the movement of the first of the two arms ( 11 ) from the first position (A) to the second position (B) is carried out along the reference path if the first passing through the step c. did not result in a collision,
ii) the movement of the first of the two arms ( 11 ) from the first position (A) to the second position (B) is carried out along an evasion path which has been determined by use of the evasion strategy, if passing through the step e. resulted in a collision.

6. Process according to claim 5 , wherein
in step ii) in addition to the movement of the first of the two arms ( 11 ) along the evasion path, the second of the two arms ( 12 ) is moved along an evasion path as well, or
in step ii) the first of the two arms ( 11 ) is moved along the reference path and the second of the two arms ( 12 ) is moved along an evasion path, or
in step ii) a movement of the first of the two arms ( 11 ) is carried out staggered in time relative to the second of the two arms ( 12 ).

7. Process according to claim 1 , wherein both arms ( 11 , 12 ) are movable simultaneously above the working area ( 10 ).

8. Process according to claim 1 , wherein when providing the actual topography, the at least one object (O 1 ) as well as other static and quasi-static objects are modelled or defined mathematically by an envelope (EH 1 , EH 2 ) in the handling system ( 100 ) and/or controller (C).

9. Process according to claim 1 , wherein when applying the evasion strategy in step f. a set of rules is applied.

10. Process according to claim 2 , wherein the set of rules comprises one or more heuristic rules.

11. Process according to claim 2 , wherein the set of rules comprises one or more of the following rules:
Determining a relative evasion height (ΔH) which is suitable for moving one of the arms ( 11 , 12 ) above one of the objects (O 1 ) without collision (Rule1);
Excluding a blocked area ( 14 ) (Rule2);
Considering of all technically possible movements of the arms ( 11 , 12 ) in the handling system ( 100 ) (Rule3);
Avoiding of pure stroke movements and/or pure lowering movements when carrying out an evasion movement (W 2 , W 3 , W 4 ) (Rule4);
Providing a security distance prior to the execution of an evasion movement (W 2 , W 3 , W 4 ) (Rule5);
Providing a smallest possible rotation if one of the arms ( 11 , 12 ) shall carry out a rotational movement (Rule6);
Examine if another than the arm addressed in the first movement request (MR 1 ) might carry out the outstanding handling or movement (Rule7 X 1 );
Examine whether the object (O 1 ) in the working area ( 10 ) may be picked up and positioned at another position (Rule7 X 2 );
Outputting a signal for advising a user of the handling system ( 100 ) of a situation which is not solvable (Rule7 X 3 );
Providing a time limit and use of another Rule (Rule8);
Initial determination (S 2 , S 3 ; S 4 . 1 ) of a path in 2-dimensions (Rule9);
Executing the determination (S 2 , S 3 ; S 4 . 1 ) of a path in a working space above the working area ( 10 ) (Rule10);
Parking of one of the arms ( 11 , 12 ) in an evasion position (Rule11);
Examine whether a movement of the first arm ( 11 ) may be carried out staggered in time relative to the second arm ( 12 ) (Rule12);
Applying a prioritization to favor one of the arms ( 11 ) over the other of the arms ( 12 ) with respect to time or path (Rule13).

12. Process according to claim 1 , wherein within the scope of one of the partial steps the image of a firmware (FW*) is used, to determine a path more accurate and to carry out the collision examination(s) of the step(s) c. and/or e. based on this more accurate path.

13. Process according to claim 12 , wherein for each of the two arms ( 11 , 12 ) an image of another firmware is used.

14. Process according to claim 1 , wherein in the scope of one of the partial steps a profile generator is used for determining a more accurate path, wherein the profile generator is also used for controlling the execution of a movement of the first of the two arms ( 11 ) from the first position (A) to the second position (B).

15. Automated handling system ( 100 ), which comprises:
a working area ( 10 ),
at least one object (O 1 ) which is placed in the working area ( 10 ),
at least a first numerically controlled arm ( 11 ),
at least a second numerically controlled arm ( 12 ), wherein these arms ( 11 , 12 ) are movable above the working area ( 10 ),
means for collision avoidance (S, MPA, CE), which are implemented in such a way that
the current actual topography of the working area ( 10 ) is determinable,
a movement planning algorithm (MPA) is applicable which determines a path from a first position (A) to a second position (B), wherein in the first application of the movement planning algorithm (MPA) a reference path, preferably the shortest path (W 1 ) is provided as a path,
a quasi-static collision examination (CE) for the path (W 1 ), which has been provided by the movement planning algorithm (MPA), is applicable,
a dynamic collision examination (CE) for the path (W 1 ), which has been suggested by the movement planning

Assignees

• Tecan Trading Ag

Inventors

• Brandenberger Joel

Classifications

• B25J9/1666

  Avoiding collision or forbidden zones · CPC title

• G01N35/0099

  comprising robots or similar manipulators (robots per se B25J) · CPC title

• G05B19/4061

  Avoiding collision or forbidden zones · CPC title

• B25J9/0087 Primary

  Dual arms (double SCARA arms B25J9/043) · CPC title

• G05B2219/40497

  Collision monitor controls planner in real time to replan if collision · CPC title