Vehicle Trajectory Collision Avoidance Using Object Sector Classification

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current vehicular navigation systems, particularly for airborne drones and autonomous vehicles, face challenges in efficiently avoiding collisions with objects due to high processing loads and the need for real-time obstacle detection and path re-planning, which can lead to delays and increased computational demands.

Innovation Solution

The system employs range finding data from sensors like lidar, radar, and sonar to classify detected objects into subsets adjacent to the vehicle's trajectory, determining whether changing direction would cause a collision, and adjusts the path accordingly, with processing either on-board or in a ground control station to minimize latency and optimize navigation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If real-time obstacle detection and path re-planning are performed using current navigation systems, then collision avoidance capability is improved, but processing load and computational demands increase significantly

Engineering Contradiction:
Improvecollision avoidance capabilityVSAvoidprocessing load
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The observation area is divided into multiple subsets (e.g., left, center, right sectors) relative to the vehicle's trajectory. Objects are classified into these subsets based on their position, allowing the system to process spatial information in manageable segments rather than as a single complex scene, thereby reducing computational load while maintaining collision avoidance capability

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Objects are pre-classified into trajectory subsets before path re-planning is initiated. This preliminary classification organizes spatial data in advance, enabling faster decision-making during critical moments when avoidance maneuvers are needed, thus reducing real-time processing demands

Inventive Principle:
Principle #10Preliminary action

2Measurement precision

If comprehensive obstacle detection is performed across the entire observation area, then detection precision is improved, but processing time increases causing delays

Engineering Contradiction:
Improveobstacle detection precisionVSAvoidprocessing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The observation area is segmented into multiple subsets, allowing the system to process spatial information in parallel across different sectors. This segmentation enables comprehensive coverage of the entire observation area while reducing the computational burden on any single processing unit, thereby maintaining detection precision without excessive processing time

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system classifies objects into trajectory subsets based on their relevance to potential collision risks. By focusing processing resources on objects within or near the trajectory subsets rather than uniformly processing all detected objects, the system achieves sufficient detection precision for safety-critical applications while reducing overall processing time

Inventive Principle:
Principle #16Partial or excessive action

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

This approach enables efficient collision avoidance with reduced processing load, allowing for real-time path adjustments and improved safety by classifying objects into sectors, validating trajectory changes, and optimizing vehicle movement to prevent collisions.

Implementation Method 1

The range finding data may be obtained using a range finding device, such as a lidar, radar, sonar or ultrasonic device

Methodology Applied
Scientific EffectRadar: Radar

Implementation Method 2

The range finding data may be obtained using a range finding device, such as a lidar, radar, sonar or ultrasonic device

Methodology Applied
Scientific EffectLIDAR: LIDAR

Implementation Method 3

The range finding data may be obtained using a range finding device, such as a lidar, radar, sonar or ultrasonic device

Methodology Applied
Scientific EffectSonar: Sonar

Data Source

PatentEP4152119B1Collision avoidance
Publication Date: 2024.06.12 NOKIA TECHNOLOGIES OY
  • EP4152119B1 patent drawingFigure 1
  • EP4152119B1 patent drawingFigure 2A~2B
  • EP4152119B1 patent drawingFigure 2C

AI summary

The present disclosure relates to vehicular navigation and collision avoidance with objects. The detected object is classified into one of the one or more subsets based on the location within the observation area. When a direction the vehicle is moving on needs to change, the detected object in a particular subset of the one or more subsets in the observation area may be taken into account in further control of the vehicle.