Intelligent Low-Cost 3D LiDAR Tripping Hazard Scanner

Date & Time

Tuesday, March 19, 2024 - 1:20 PM PDT  

Location

Platinum 1  

Description

A high-resolution 3D LiDAR scanner was developed to detect tripping hazards with small elevation changes, such as potholes, doorsteps or curbs. Although these obstacles are rarely a problem for people with normal vision, they are the most common tripping hazards for blind and visually impaired people. These small elevation differences can blend with the background road surface, making them difficult to detect by sonar, 2D LiDAR or camera. This paper introduces a new way to turn a low-cost 2D LiDAR into a high-resolution 3D LiDAR and generate a live 3D terrain model that can effectively identify elevation changes as small as 2cm. The feasibility of this 3D sensor has been demonstrated to reliably identify potholes with a minimum size of 2cm(W)x7cm(L)x2cm(D) at a distance up to 2m ahead of the traveler and operational up to 8m (>4x the range of traditional white canes).

Inspired by autonomous driving and robotic vision technologies, the ground/floor in the 3D terrain model was removed using Random Sample Consensus (RANSAC) to create segmented point clouds of the tripping hazards, followed by finding the Oriented Bounding Box (OBB) using Convex Hull and Rotating Calipers techniques. In this way, each obstacle is not only isolated from the background with a clear location marked but also the orientation and dimensions are quantified in 3D using an OBB.

This Presentation Link is provided by the Presenter(s) and not hosted by the Center on Disabilities at CSUN. The Center on Disabilities has confirmed, as of March 27, 2024, content linked is relevant to the presentation, but has not been reviewed for accessibility nor will the Center on Disabilities attempt to remediate any accessibility issues in the linked content. Please contact the Presenter(s) with any accessibility concerns.

 

Audience

• Higher Education
• Information & Communications Technology
• Disability Specific
• Research & Development
• Transportation

Audience Level

Intermediate  

Session Summary (Abstract)

A low-cost 3D LiDAR-based tripping hazard detector is introduced to differentiate elevation changes as small as 2cm, combined with an intelligent computer vision algorithm to track multiple obstacles’ locations and dimensions.  

Primary Topic

Emerging Technologies  

Secondary Topics

• Artificial Intelligence (AI) & Machine Learning (ML)
• Blind/Low Vision
• Engineering
• Research

 

Session Type

Journal Track  

Presenter

• Brandon Cai
  Stony Brook University

Add to Outlook/iCal (How it works)

Back to Session List