Bird's-Eye Obstacle Mapping for Small Object Avoidance

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Autonomous moving bodies face challenges in detecting and avoiding obstacles like dirt and small falling objects on a road surface, as existing sensors may not recognize these, leaving the operation to user intervention which is difficult to manage effectively.

Innovation Solution

An information processing device that acquires images, recognizes attributes, and generates a bird's-eye view map integrating texture and semantic maps to provide users with a clear visual representation of the environment, enabling easier instruction and route adjustment for the moving body.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If sensors are used to detect obstacles, then the autonomous moving body can avoid recognized objects, but small objects like dirt and falling objects cannot be detected

Engineering Contradiction:
Improveobstacle detection capabilityVSAvoiddetection of small objects
Core Design Contradiction:
ReliabilityVSDifficulty of detecting and measuring

Solution Approach 1:

The patent introduces an intermediary information processing system that captures images, recognizes object attributes, and generates a bird's-eye view map with semantic information. This intermediary system bridges the gap between sensor detection and user decision-making, allowing users to see and instruct avoidance of small objects that sensors cannot detect

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The system provides visual feedback through a bird's-eye view map displaying recognized objects and their attributes. This feedback loop allows users to understand the situation and provide instructions for avoiding small objects, transforming the autonomous moving body into a semi-autonomous system with user oversight

Inventive Principle:
Principle #23Feedback

2Reliability

If user intervention is used to avoid undetected objects, then small objects can be avoided, but it becomes difficult to grasp the situation for instructing operation

Engineering Contradiction:
Improveavoidance of small objectsVSAvoiduser instruction capability
Core Design Contradiction:
ReliabilityVSEase of operation

Solution Approach 1:

The patent transforms the two-dimensional image data into a three-dimensional bird's-eye view map with additional semantic dimensions. This dimensional transformation provides users with comprehensive spatial and attribute information, making it easy to grasp the situation and provide accurate avoidance instructions

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

Solution Approach 2:

The system creates a simplified copy of the real-world environment in the form of a bird's-eye view map with recognized objects and their attributes. This information copy allows users to easily understand the situation without directly observing the complex real-world scene

Inventive Principle:
Principle #26Copying

Data Source

PatentUS11812197B2Information processing device, information processing method, and moving body
Publication Date: 2023.11.07 SONY GROUP CORP
  • US11812197B2 patent drawing
  • US11812197B2 patent drawing
  • US11812197B2 patent drawing

AI summary

Provided is an information processing device that includes an acquisition unit that acquires an image captured by an imaging unit, a recognition unit that recognizes attributes of an object shown in the image captured by the imaging unit, and a generation unit that generates a bird's-eye view map showing the attributes of the object on the basis of the image captured by the imaging unit and information on the attributes of the object recognized by the recognition unit.