Work Vehicle Trajectory Search Using Route-Guided Penalty Constraints

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing automatic driving systems for work vehicles face challenges in calculating a continuous travel track due to high nonlinearity and complexity of obstacle restrictions, leading to increased calculation amounts and potential convergence to local solutions.

Innovation Solution

A two-stage track search method using a route search unit and a track search unit, where the route search unit determines a series of positions and postures based on a first restriction condition, and the track search unit determines a series of positions, postures, speeds, and steering angles based on a penalty value-increasing second restriction condition, reducing calculation complexity and ensuring continuity.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If a mathematical model with multiple dimensions (position, posture, steering angle, speed, acceleration) is used to search for a track, then the accuracy of the track is improved, but the calculation amount increases and convergence to local solutions occurs

Engineering Contradiction:
Improvetrack accuracyVSAvoidcalculation complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the track search into two distinct stages: route search (coarse-level path planning) and track search (fine-level trajectory optimization). The route search first determines a rough path avoiding obstacles, and then the track search refines this path into a smooth, continuous trajectory with proper kinematic constraints. This segmentation reduces the overall calculation complexity while maintaining track accuracy.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The route search is performed as a preliminary action before the track search. By first establishing a rough route that satisfies obstacle avoidance constraints, the subsequent track search only needs to refine the trajectory along this predetermined path, significantly reducing the search space and calculation burden while ensuring the final track maintains accuracy.

Inventive Principle:
Principle #10Preliminary action

2Device complexity

If the number of dimensions of a model is reduced to reduce the amount of calculation, then the calculation amount is reduced, but operations related to the reduced dimensions may be discontinuous in an obtained track

Engineering Contradiction:
Improvecalculation complexityVSAvoidtrack continuity
Core Design Contradiction:
Device complexityVSStability of the object's composition

Solution Approach 1:

The patent applies dynamic programming in the track search stage to dynamically optimize the trajectory while maintaining continuity. The dynamic programming approach systematically explores the state space (position, posture, speed, steering angle) and ensures that the optimized track maintains smooth transitions and continuous operations, avoiding discontinuities that would result from simple dimension reduction.

Inventive Principle:
Principle #15Dynamics

3Reliability

If obstacle positions are used as restriction conditions in track search, then obstacle avoidance is achieved, but the nonlinearity of restriction conditions increases making on-line calculation difficult

Engineering Contradiction:
Improveobstacle avoidanceVSAvoidrestriction condition complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The route search is performed as a preliminary action that pre-processes the obstacle avoidance constraints. By first determining a route that satisfies obstacle avoidance requirements, the complex nonlinear restriction conditions are transformed into a simpler framework for the subsequent track search, making on-line calculation feasible while maintaining reliable obstacle avoidance.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS12461542B2System, method, and work vehicle
Publication Date: 2025.11.04 KOMATSU LTD
  • US12461542B2 patent drawing
  • US12461542B2 patent drawing
  • US12461542B2 patent drawing

AI summary

A system searches for a track for a vehicle to travel automatically. The system includes a processor. The processor includes a route search unit, a restriction condition generation unit, and a track search unit. The route search unit searches for a series having elements of a position and posture of the vehicle, which is a route for moving from an initial position to a target position of the vehicle, based on a first restriction condition representing a position of an obstacle. The restriction condition generation unit generates a second restriction condition in which a penalty value increases according to a deviation distance from the route. The track search unit searches for a series having elements of a position, posture, speed, and steering angle of the vehicle, which is a track for moving from the initial position to the target position of the vehicle, based on the second restriction condition.