IP Library Granted Patent US 12703364
Granted Patent B2
US 12703364 · App. 18/748,299 · Granted Aug 11, 2026

Autonomous control system and autonomous control method

Inventors: Shinji Ishihara (Tokyo, JP); Masaki Kanai (Tokyo, JP); Teppei Saitoh (Tokyo, JP); Ryu Narikawa (Tokyo, JP); Kazuya Sugimoto (Tokyo, JP); Tadashi Kotani (Tokyo, JP)
Assignee: HITACHI, LTD.
B60W30/18154B60W60/001B60W2554/4029B60W2554/4041
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12703364
App. No.
18/748,299
Granted
Aug 11, 2026
Kind
B2
Abstract

An autonomous control system includes a state amount acquiring unit that acquires the information of a control target, a control target position identification unit, a control target reaching region calculation unit, an outside information acquiring unit, a target distinguishing unit, a non-control target position identification unit, a non-control target reaching region calculation unit, an intersection region calculation unit, a control parameter setting unit, and an operation amount calculation unit that calculates a control input such that the value of an objective function J becomes a smaller value than the previous value.

Claims (46)

1 . An autonomous control system comprising:

a state amount acquiring unit that acquires a state amount of a control target;

a control target position identification unit that identifies a position of the control target on the basis of the state amount;

a control target reaching region calculation unit that calculates a control target reaching region that the control target can reach within a predetermined time on the basis of the position of the control target;

an outside information acquiring unit that acquires outside information of the control target;

a target distinguishing unit that distinguishes an attribute of a non-control target on the basis of the outside information;

a non-control target position identification unit that calculates a position of the non-control target on the basis of the outside information;

a non-control target reaching region calculation unit that uses the attribute and the position, to calculate a non-control target region that the non-control target can reach within a predetermined time, wherein the attribute is used to classify a movement potential of the non-control target for calculating the non-control target region based at least on holonomic constraints associated with each different classification;

an intersection region calculation unit that uses a set of the positions of the control target and the set of the positions of the non-control target calculated by the non-control target reaching region calculation unit, to calculate an intersection region of the sets of the respective positions;

a control parameter setting unit that sets a control parameter of the control target such that as an intersection amount region of the intersection region of the sets of the respective positions is larger, a behavior in which the control target widens the distance between the control target and the non-control target is generated; and

an operation amount calculation unit that uses the control parameter and an objective function whose value becomes smaller as the control target is closer to the desirable behavior, to calculate a control input such that the value of the objective function becomes a smaller value than the previous value in the range in which a predetermined constraint condition is satisfied.

2 . The autonomous control system according to claim 1 ,

wherein the non-control target reaching region calculation unit uses the attribute distinguished by the target distinguishing unit and the position of the non-control target calculated by the non-control target position identification unit, to calculate a predictive trajectory as a set of trajectories through which the non-control target passes within the predetermined time and the non-control target region that the non-control target can reach within the predetermined time,

wherein the intersection region calculation unit calculates a first intersection region by using the set of the positions of the control target calculated by the control target reaching region calculation unit and the set of the positions of the non-control target calculated by the non-control target reaching region calculation unit, and calculates a second intersection region by using the set of the positions of the control target calculated by the control target reaching region calculation unit and the set of the predictive trajectories calculated by the non-control target reaching region calculation unit, and

wherein the control parameter setting unit sets the control parameter such that even in a situation where the values of the set of the first intersection region are the same value, as the set of the second intersection region is larger, the behavior in which the distance between the control target and the non-control target is widened is generated.

3 . The autonomous control system according to claim 1 ,

wherein the information distinguished by the target distinguishing unit includes, in addition to the movement potential of the non-control target, a movement characteristic regarding safe movement that can be taken by the non-control target, and

wherein the control parameter setting unit sets the control parameter such that even when the values of the set of the intersection region calculated by the intersection region calculation unit are the same, as the non-control target has the movement characteristic in which the non-control target is less likely to take the safe movement, the behavior in which the distance between the control target and the non-control target is widened is generated.

4 . The autonomous control system according to claim 1 ,

wherein the intersection region calculation unit uses the set of the positions at respective times of the control target calculated by the control target reaching region calculation unit and the set of the positions at respective times of the non-control target calculated by the non-control target reaching region calculation unit, to calculate the intersection region of the sets of the respective positions at the respective times, and

wherein the control parameter setting unit sets the control parameter such that even when the values of the set of the intersection region calculated by the intersection region calculation unit are the same values, as the values of the set of the intersection region at the same time of the set of the positions at the respective times of the control target and the set of the positions at the respective times of the non-control target are larger, the behavior in which the distance between the control target and the non-control target is widened is generated.

5 . The autonomous control system according to claim 1 ,

wherein the objective function that is used in the operation amount calculation unit includes the product of a penalty function whose value becomes larger as the distance between the control target and the non-control target is smaller and the control parameter decided by the control parameter setting unit.

6 . The autonomous control system according to claim 1 ,

wherein the constraint condition includes a minimum distance in which the control target and the non-control target can be closer to each other, and for the minimum distance, the control parameter decided by the control parameter setting unit is used.

7 . The autonomous control system according to claim 1 ,

wherein the control target is a vehicle.

8 . The autonomous control system according to claim 1 ,

wherein the control target is a hydraulic excavator.

9 . The autonomous control system according to claim 1 ,

wherein the control target is a flying body.

10 . An autonomous control method comprising:

acquiring a state amount of a control target;

identifying a position of the control target on the basis of the state amount;

calculating a control target reaching region that the control target can reach within a predetermined time on the basis of the position of the control target;

acquiring outside information of the control target;

distinguishing an attribute of a non-control target on the basis of the outside information;

calculating a position of the non-control target on the basis of the outside information;

using the attribute and the position, calculating a non-control target region that the non-control target can reach within a predetermined time, wherein the attribute is used to classify a movement potential of the non-control target for calculating the non-control target region based at least on holonomic constraints associated with each different classification;

using a set of the positions of the control target and a set of the positions of the non-control target, calculating an intersection region of the sets of the respective positions;

setting a control parameter of the control target such that as an intersection amount region of the intersection region of the sets of the respective positions is larger, a behavior in which the control target widens the distance between the control target and the non-control target is generated; and

using the control parameter and an objective function whose value becomes smaller as the control target is closer to the desirable behavior, calculating a control input such that the value of the objective function becomes a smaller value than the previous value in the range in which a predetermined constraint condition is satisfied.

11 . The autonomous control method according to claim 10 , comprising:

using the attribute and the position of the non-control target, calculating a predictive trajectory as a set of trajectories through which the non-control target passes within the predetermined time and the non-control target region that the non-control target can reach within the predetermined time;

calculating a first intersection region by using the set of the positions of the control target and the set of the positions of the non-control target, and calculating a second intersection region by using the set of the positions of the control target and the set of the predictive trajectories; and

setting the control parameter such that even in a situation where the values of the set of the first intersection region are the same value, as the set of the second intersection region is larger, the behavior in which the distance between the control target and the non-control target is widened is generated.