Bystander Pose Estimation for Material Handling Vehicle Navigation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Material handling vehicles (MHVs) face challenges in navigating environments safely and efficiently, particularly in detecting bystanders and adjusting their operations to avoid collisions, due to limitations in existing sensor technologies and training requirements for operators.

Innovation Solution

An autonomous bystander control system using a machine learning-based automation processing system that integrates sensor data from various types, such as cameras and LIDAR, to detect and estimate bystander presence and pose, and generate control actions for the MHV, enabling proactive navigation and communication with other MHVs to adjust routes.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If traditional sensor technologies are used for MHV navigation, then the system complexity is lower, but the ability to detect and respond to bystanders is insufficient

Engineering Contradiction:
Improvebystander detection capabilityVSAvoidsystem complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The system segments bystander detection into multiple specialized components: a bystander detection model for identifying presence, a pose estimation model for determining body orientation, and a hand signal recognition model for interpreting gestures. Each component processes specific aspects of bystander information independently, then integrates results to achieve comprehensive detection capability without overwhelming system complexity

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system transitions from traditional 2D image processing to 3D pose estimation by inferring three-dimensional body positions and orientations from sensor data. This dimensional enhancement enables the MHV to understand spatial relationships and bystander intentions more accurately, improving detection reliability while managing complexity through algorithmic efficiency

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

2Reliability

If operator training and supervision are implemented to ensure safety, then collision risk is reduced, but operational efficiency and productivity decrease

Engineering Contradiction:
ImprovesafetyVSAvoidoperational efficiency
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The MHV system performs self-monitoring for safety by autonomously detecting bystanders, estimating their poses, and interpreting hand signals to make real-time navigation decisions. This self-service safety mechanism eliminates the need for continuous human supervision and training interventions, maintaining high safety standards while preserving operational efficiency and productivity

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The system implements continuous feedback loops where sensor data from cameras and LIDAR is processed to detect bystanders, estimate poses, and recognize hand signals. This real-time feedback enables the MHV to dynamically adjust its navigation path and speed based on detected conditions, ensuring safety without requiring human intervention that would reduce productivity

Inventive Principle:
Principle #23Feedback

3Measurement precision

If multiple sensors are integrated for comprehensive environment perception, then detection accuracy improves, but device complexity and cost increase

Engineering Contradiction:
Improvedetection accuracyVSAvoidsensor integration complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The system merges data from multiple sensor types including cameras and LIDAR into a unified processing framework. By combining visual imagery with depth information from LIDAR, the system achieves accurate 3D pose estimation and hand signal recognition. The merging is managed through integrated processing that handles multi-sensor data streams efficiently, improving detection accuracy while controlling complexity through unified architecture

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The sensor system is designed with multi-functionality where the same sensor array serves multiple purposes: detecting bystander presence, estimating body pose, recognizing hand signals, and mapping the environment. This universal sensor platform improves detection accuracy across all functions while reducing overall system complexity compared to having separate specialized sensors for each task

Inventive Principle:
Principle #6Universality (Multi-functionality)

Data Source

PatentEP4369136A1Systems and methods for bystander pose estimation for industrial vehicles
Publication Date: 2024.05.15 RAYMOND LTD
  • EP4369136A1 patent drawingFigure 1
  • EP4369136A1 patent drawingFigure 2A~2B
  • EP4369136A1 patent drawingFigure 3

AI summary

Systems and methods for enhanced MHV operation using an automation processing system for bystander detection and bystander pose estimation to control operation of the MHV.