Automate AutoCAD Linework & Point Creation from LIDAR

A while back, I was working on a project when we realized we needed one more piece of existing site information.

Nothing major.

Just a feature that hadn’t been picked up clearly enough during the original survey.

The problem?

The survey had been done weeks ago.

If you’ve worked in civil, surveying, architecture, or engineering long enough, you’ve probably dealt with some version of this.

And it highlights a bigger problem with the traditional workflow:

We only bring back the information we thought we’d need at the time.

But reality capture has changed this quite a bit.

Instead of individually shooting every curb, ditch, tree, utility pole, fence post, sign, and spot elevation, we can capture an incredibly detailed digital record of the site once...

Then extract the information we need later.

And with newer “digital surveying” tools, we’re getting much better at turning that massive amount of reality data into something we can actually use in CAD.

One of the easiest ways to create a messy base plan is importing survey or consultant data directly into your working drawing without cleaning it first.

Instead, create a dedicated Survey / Existing Conditions Xref and standardize it before using it throughout your project.

A few things I like to check:

• Layer names and properties

• Units and coordinate system

• Duplicate or overlapping geometry

• Unnecessary points and objects

• Z elevations

• Linetypes

• Blocks and symbols

If the incoming layers don’t match your standards, LAYTRANS is a great tool for mapping them to your existing layer structure.

Clean it once.

Xref it into the project.

Then every drawing uses the same organized source of truth.

We’ve gotten incredibly good at collecting reality data.

Drones, mobile mapping systems, and handheld LiDAR scanners can capture millions of points across a project site incredibly quickly.

But there’s still a problem...

Someone has to turn all of that data into a usable drawing.

That’s what I wanted to explore in this week’s video.

I’m testing Mach9 Digital Surveyor, which is designed to automate and streamline the process between:

LiDAR Scan → Survey Data → CAD Drawing

Using a real roadway scan from San Francisco, I walk through how Digital Surveyor can automatically identify and generate a large portion of the existing site linework.

Curbs
Road edges
Markings
Trees
Poles
Signs
And more

It can even automatically organize those features onto the correct CAD layers using predefined or custom layer templates.

From there, I test some of its semi-automated surveying tools.

The Linear Extraction Tool helps build and refine curb, gutter, roadway, and other linear features without manually clicking every vertex.

A Cross Section Tool lets you slice through the point cloud and quickly QA/QC the extracted geometry.

And the Grid of Points Tool lets you generate useful terrain and survey points from the scan without bringing millions of unnecessary LiDAR points into Civil 3D.

Once everything is cleaned up and verified...

We export the finished survey directly to a DWG and open it in AutoCAD and Civil 3D.

That’s the part of this workflow I find most interesting.

We’re not just collecting more data.

We’re getting much better at turning that data into something designers can actually use.

Watch the full video here:
Create Survey Plans from LIDAR Fast!

What it does:
Turns LiDAR and reality-capture datasets into usable CAD, GIS, and survey information.

Why it’s interesting:
Digital Surveyor combines automatic feature extraction with semi-automated CAD and surveying tools.

Instead of manually tracing an entire point cloud, much of the initial linework can be generated automatically.

You can then review it, refine missing features, create additional survey points, verify geometry using cross sections, and export the finished result to DWG.

One feature I especially like is the ability to create or import your own layer templates.

That means the software can identify a curb, tree, pole, road edge, or other feature and automatically place it onto the layer that matches your existing CAD standards.

The end result is a workflow that looks something like:

Scan the Site → Automatically Extract Features → Digitally Survey → QA/QC → Export DWG → Start Designing

And I think that’s where reality capture becomes really powerful.

You’re no longer limited to only the individual features someone thought to survey that day.

If you need another curb line, terrain point, or roadway feature later in the project, the information may already exist in your scan.

You just need to extract it.

How are you using and processing Lidar/Reality Capture Data?

Hit reply and let me know I’m always looking for new tools and workflows to test and share.

P.S. If you’ve got a colleague who’s excited about LIDAR and scanning, be sure to forward them this email or send them a link!

They’ll thank you later.