METHOD FOR PLANNING PATH FOR AUTONOMOUS WALKING HUMANOID ROBOT

• US 20140025201A1
• Filed: 12/31/2012
• Published: 01/23/2014
• Est. Priority Date: 07/19/2012
• Status: Active Grant

• Current Assignee: Center of Human-centered Interaction for Coexistence (Government of South Korea)
• Original Assignee: Center of Human-centered Interaction for Coexistence (Government of South Korea), Korea University Research and Business Foundation (Korea University)
• Inventors: Doh, Nak Ju, You, Bum Jae, Ryu, Soo Hyun, Kang, Yeon Sik
• Granted Patent: US 9,069,354 B2
• Time in Patent Office: Days
• US Class Current: 700/245
• CPC Class Codes: G05D 1/02 Control of position or course in two dimensions
• CPC Class Codes: G05D 1/0214 in accordance with safety or protection criteria, e.g. avoiding hazardous areas monitoring the location of vehicles within a certain area, e.g. forbidden or allowed areas, in traffic control systems for road vehicles G08G1/13
• CPC Class Codes: G05D 1/0217 in accordance with energy consumption, time reduction or distance reduction criteria
• CPC Class Codes: G05D 1/024 in combination with a laser lasers per se H01S
• CPC Class Codes: G05D 1/0274 using mapping information stored in a memory device navigation using map-matching G01C21/30
• CPC Class Codes: Y10S 901/01 Mobile robot

Abstract

The present invention provides a method for planning a path for an autonomous walking humanoid robot that takes an autonomous walking step using environment map information, the method comprising: an initialization step of initializing path input information of the autonomous walking humanoid robot using origin information, destination information, and the environment map information; an input information conversion step of forming a virtual robot including information on the virtual robot obtained by considering the radius and the radius of gyration of the autonomous walking humanoid robot based on the initialized path input information; a path generation step of generating a path of the virtual robot using the virtual robot information, the origin information S, the destination information G, and the environment map information; and an output information conversion step of converting the path of the autonomous walking humanoid robot based on the virtual robot path generated in the path generation step.

Claims

1. A method for planning a path for an autonomous walking humanoid robot that takes an autonomous walking step using environment map information, the method comprising:

an initialization step of initializing path input information of the autonomous walking humanoid robot using origin information, destination information, and the environment map information;

an input information conversion step of forming a virtual robot including information on the virtual robot obtained by considering the radius and the radius of gyration of the autonomous walking humanoid robot based on the initialized path input information;

a path generation step of generating a path of the virtual robot using the virtual robot information, the origin information S, the destination information G, and the environment map information; and

an output information conversion step of converting the path of the autonomous walking humanoid robot based on the virtual robot path generated in the path generation step.

2. The method according to claim 1, wherein the input information conversion step comprises:

a virtual robot formation step of including virtual robot information containing a radius of the virtual robot as a value obtained by adding the radius of the autonomous walking humanoid robot and the radius of gyration of the autonomous walking humanoid robot extending from the center of the autonomous walking humanoid robot to the center of gyration of the autonomous walking humanoid robot; and

an obstacle extension region formation of forming an obstacle extension region of the obstacle based on the environment map information.

3. The method according to claim 1, wherein the path generation step comprises:

a macro path generation step of calculating a macro path candidate group from the origin information to the destination information, calculating the time cost required for the autonomous walking humanoid robot to travel along the calculated macro path candidate group, and calculating a macro path having the minimum time cost among the time costs for the macro path candidate group; and

a micro path control step of controlling the macro path by calculating an alternative path passing through an intersecting point on path extension lines using the macro path calculated in the macro path generation step and comparing the respective time costs.

4. The method according to claim 3, wherein the macro path generation step comprises:

a macro path candidate group formation step of forming the macro path candidate group as paths passing through the outside of the obstacle extension region extending from the origin information to the destination information using the origin information, the destination information, the environment map information, and the obstacle extension region;

a macro path candidate group time cost calculation step of calculating the time costs for all the macro path candidate of the macro path candidate group using a walking step pattern (s) on a path between two nodes formed by each macro path candidate of the macro path candidate group, a distance (l) of translation on the path between the nodes, and node rotation angle information (θ

tot=θ

+Δ

θ

) determined by considering the forward and backward walking step patterns for the nodes; and

a macro path selection step of comparing the time costs calculated in the macro path candidate group time cost calculation step and selecting a macro path candidate having a minimum time cost among the compared time costs as a final macro path.

5. The method according to claim 4, wherein the node rotation angle information (θ

tot=θ

+Δ

θ

) comprises;

a node rotation reference angle information θ

determined when a walking step pattern (sprev) of the autonomous walking humanoid robot on a path between the node on the macro path candidate and a previous node is a forward step (FS) and a walking step pattern (snext) of the autonomous walking humanoid robot on a path between the node and a next node is a forward step (FS); and

a node rotation step angle Δ

θ

determined depending on the walking step pattern (sprev) of the autonomous walking humanoid robot on the path between the node on the macro path candidate and the previous node and the walking step pattern (snext) of the autonomous walking humanoid robot on the path between the node and the next node.

6. The method according to claim 3, wherein the micro path control step comprises:

an edge selection step of selecting three consecutive unit edges on the macro path;

a path extension line calculation step of calculating path extension lines extending on the macro path for the left and right outermost edges among the three consecutive edges selected in the edge selection step;

an intersecting point calculation step of calculating an intersecting point between the path extension lines;

an alternative path calculation step of calculating an alternative path formed by the intersecting point and the left and right outermost edges;

a path cost calculation step of calculating the time costs for the alternative path and the macro path including the three consecutive edges selected in the edge selection step; and

a controlled path selection step of comparing the time costs for the alternative path and the macro path including the three consecutive edges selected in the edge selection step with each other and selecting a macro path having a minimum time cost among the compared time costs as a controlled path, and setting the controlled path as a path of the virtual robot for the selected edges.