Autonomous mobile machine, and system and method for controlling a mobile machine

• US 5,545,960 A
• Filed: 04/06/1992
• Issued: 08/13/1996
• Est. Priority Date: 04/09/1991
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Ishikawa, Shigeki
• Primary Examiner(s): IP, SHIK LUEN PAUL
• Application Number:
  US07/865,576
• Time in Patent Office: 1,590 Days
• Field of Search: 318/560-646, 318/139, 364/424.02, 364/424.06, 364/426, 364/443, 395/80-89, 901/1, 901/3, 901/9, 901/12, 901/13, 901/15-20, 180/167-169
• US Class Current: 318/587
• CPC Class Codes: G05D 1/0242 using non-visible light signals, e.g. IR or UV signals
• CPC Class Codes: G05D 1/0255 using acoustic signals, e.g. ultra-sonic singals sonar systems designed for anti-collision purposes G01S15/93
• CPC Class Codes: G05D 1/0272 comprising means for registering the travel distance, e.g. revolutions of wheels measuring distance traversed on the ground by vehicles, e.g. using odometers G01C22/00
• CPC Class Codes: G05D 1/0278 using satellite positioning signals, e.g. GPS

Abstract

The present apparatus, control system, and method for navigating is used to navigate a mobile machine accurately along a traced path so that the machine avoids collisions with obstacles. Intermediate control outputs are generated not only in accordance with a signal indicative of a distance and angle from a path to be traced according to a navigation procedure for path tracing, but also in response to a signal related to the obstacle-free space ahead of the machine according to a navigation procedure for obstacle avoidance. Weighting coefficients respective for the two procedures are generated by fuzzy inference based on signals indicating the distance and angle from the predetermined path and signals related to the obstacle-free space ahead of the machine. Final control outputs are computed in accordance with the two intermediate outputs and the two weighting coefficients. Regardless of the number of obstacles a mobile machine comes across while moving along a path, and of whether or not they are moving, the mobile machine can avoid them reliably and smoothly.

Claims

1. An autonomous mobile machine comprising:

(a) a sensing unit for outputting signals related to a distance and an angle between the machine and a path to be traced the sensing unit further outputting signals related to an obstacle-free space, the obstacle-free space being the space in which the machine can move freely between a first and second sampling time;

(b) means for obtaining intermediate control outputs by using a navigation procedure for path tracing in accordance with signals related to the distance and angle;

(c) means for obtaining intermediate control outputs by using a navigation procedure for obstacle avoidance in accordance with signals related to the obstacle-free space;

(d) means for determining a first weighting coefficient for path tracing and a second weighting coefficient for obstacle avoidance by inference based on a set of fuzzy rules in accordance with the signals related to the distance and angle and the signals related the obstacle-free space;

(e) means for computing a set of computed control outputs from the intermediate outputs obtained by said means (c), those obtained by the means (d), a first coefficient, and a second coefficient; and

(f) a drive system for driving the machine in accordance with the set of computed control outputs.

2. The autonomous mobile machine according to claim 1 wherein the means (c) performs inference by using an obstacle set of fuzzy rules, the obstacle set being a subset of the set of fuzzy rules that are related to obstacle avoidance.

3. The autonomous mobile machine according to claim 2 wherein the set of fuzzy rules for determining the weighting coefficients and the obstacle set of fuzzy rules for obstacle avoidance have a one-to-one correspondence, and wherein corresponding rules have the same condition.

4. The autonomous mobile machine according to claim 1 wherein the signals related to obstacle-free space consist of a signal indicating the distance from the machine to the obstacle in each of a plurality of spaces called divisional spaces determined by dividing the space ahead of the current position of the machine into one or more spaces, and a signal indicating change in a free space distance for each divisional space, the free space distance being the distance between the machine and the object in each respective divisional space between sampling times.

5. The autonomous mobile machine according to claim 1 wherein the fuzzy values of the second weighting coefficient are described in the set of fuzzy rules.

6. The autonomous mobile machine according to claim 5 wherein the set of fuzzy rules includes rules that output a large fuzzy value for the second weighing coefficient in response to a distance fuzzy value of the free space distance to the obstacle being small in all of the divisional spaces.

7. The autonomous mobile machine according to claim 5 wherein said set of fuzzy rules includes rules that output a large fuzzy value for the second weighing coefficient in response to a fuzzy rule value of the free space distance to an obstacle being small in one of a plurality of divisional spaces located ahead of the machine with respect to the machine'"'"'s current direction of motion called a front space.

8. The autonomous mobile machine according to claim 5 wherein the set of fuzzy rules includes rules that output a small fuzzy value for the second weighting coefficient in response to a distance fuzzy value of the free space distance to an obstacle being large in a front space.

9. The autonomous mobile machine according to claim 5 wherein the set of rules includes rules that output a zero fuzzy value for the second coefficient in response to a distance change fuzzy value of the change in the distance to the obstacle between sampling times being zero and a fuzzy value indicating that the distance and angle from the path being zero in all of a plurality of divisional spaces.

10. The autonomous mobile machine according to claim 5 wherein the set of fuzzy rules includes rules that output a medium fuzzy value for the second weighting coefficient in response to a distance fuzzy value of the divisional distance to the obstacle being large in a front space but small in at least one of a plurality of remaining divisional spaces.

11. The autonomous mobile machine according to claim 5 wherein the set of fuzzy rules includes rules that output a medium fuzzy value for the second weighting coefficient in response to a distance fuzzy value of the divisional distance to the obstacle being large in a front space but a distance change fuzzy value of the change in this distance between sampling times being negative.

12. The autonomous mobile machine according to claim 5 wherein the set of fuzzy rules includes rules that output a large fuzzy value for the second weighting coefficient in response to a distance fuzzy value of the distance to the obstacle located to one side of the machine'"'"'s current direction of motion being zero and a fuzzy value indicating that the distance and angle from the path is large.

13. The autonomous mobile machine according to claim 1 including means for generating a STATE and a SIDE signal the SIDE signal indicating the stationary degree of the machine'"'"'s path tracing and the STATE signal indicating the dynamic degree of the machine'"'"'s path tracing.

14. The autonomous mobile machine according to claim 9 wherein a fuzzy value indicating the distance and angle from the path as a fuzzy value of a dynamic degree of path tracing.

15. The autonomous mobile machine according to claim 12 wherein a fuzzy value of the distance and angle from the path is a fuzzy value of a stationary degree of path tracing.

19. The autonomous mobile machine according to claim 4 wherein the fuzzy values of the second weighting coefficient are described in the set of fuzzy rules.

16. A control apparatus for navigating a mobile machine comprising:

means for obtaining intermediate control outputs in accordance with signals related to a distance and an angle between the machine and a path to be traced according to a navigation procedure for path tracing;

means for obtaining intermediate control outputs in accordance with signals related to an obstacle-free space according to the navigation procedure for obstacle avoidance;

means for determining a first weighting coefficient for path tracing and a second weighting coefficient for obstacle avoidance by inference based on a set of fuzzy rules in response to the signals related to distance and angle and the signals related to obstacle-free space; and

means for computing control outputs from a set of intermediate outputs obtained by the navigation procedure for path tracing, those obtained by the navigation procedure for obstacle avoidance, the first coefficient, and the second coefficient.

17. A control method for navigating a mobile machine comprising the steps of:

obtaining intermediate control outputs in accordance with information related to a distance and one or more angles between the machine and a path to be traced according to a navigation procedure for path tracing;

obtaining intermediate control outputs in accordance with information related to an obstacle-free space according to the navigation procedure for obstacle avoidance;

determining a first weighting coefficient for path tracing and a second weighting coefficient for obstacle avoidance in accordance with said information related to the distance and angle and said information related to the obstacle-free space; and

computing control outputs in accordance with the intermediate outputs obtained by said navigation procedure for path tracing, the first weighting coefficient, the intermediate outputs obtained by said navigation procedure for obstacle avoidance, and the second weighting coefficient.

18. A control system comprising:

a sensing unit for outputting different kinds of signals indicating one or more control parameters for calculating the weighting coefficients;

means for generating intermediate control outputs according to a first control procedure in accordance with at least one kind of signal;

means for generating intermediate control outputs according to a second control procedure in accordance with at least one other kind of signal;

means for generating a first weighting coefficient for the first control procedure and a second weighting coefficient for the second control procedure by fuzzy inference in accordance with the kinds of signals used in the first and second control procedures; and

means for computing control outputs in accordance with the intermediate outputs obtained by first control procedure, the first weighting coefficient, the intermediate outputs obtained by second control procedure, and the second weighting coefficient.