logoMain Street Autonomy

Why Calibrate?

Correct sensor calibration is essential to a successful robot deployment.

chevron_right Fleets of robots all behave similarly. No ‘gold units’, no ‘duds’
chevron_right Robot maintenance is easier, faster, and less expensive
chevron_right ML systems generalize better with calibrated sensor data

Calibration makes sensor data more usable


chevron_right Accuracy: Calibration adjusts sensor outputs to more closely match the real world, reducing errors and improving accuracy
chevron_right Precision: Calibration ensures similar sensors generate similar results, improving consistency and simplifying maintenance
chevron_right Sensor fusion: Calibration enables different sensor modalities to perceive the world in the same way, producing coherent and correctly-aligned data that can be fused together

Calibration Anywhere

MSA’s automatic sensor calibration solution works with any number, combination, and layout of perception sensors in any unstructured environment. Calibration Anywhere generates sensor intrinsics, extrinsics, and time offsets for all perception sensors in one pass.

check_box Calibrate anywhere - no checkerboards, no targets, no special environment
check_box Typically takes less than 10 minutes
check_box No engineers or technicians are required

Supported Sensors

Any number of perception sensors with almost any layout and configuration:

• Cameras
• RGB & Thermal
• Global and rolling shutter
• Stereo and ToF depth
• Lidars
• 2D and 3D
• Radars
• IMUs
• GPS/GNSS units
• Wheel encoders

Calibrated Outputs

Generated in one pass in typically less than 10 minutes:

• Extrinsics relative to base_link coordinate frame (including an NVIDIA Isaac Perceptor compatible URDF)
• 6DoF extrinsics for cameras, 3D lidars, radars, IMUs
• 3DoF extrinsics for 2D lidars
• 3D position of GPS/GNSS units
• Camera lens intrinsics for cameras
• Supporting fisheye, equidistant, ftheta3, rational polynomial, and plumbob models
• Includes readout time for rolling shutter cameras
• Time offsets (timestamp corrections)
• For cameras, lidars, radars, IMUs, wheel encoders, and GPS/GNSS units
• Ground detection
• Ground plane relative to cameras, 3D lidars, and base_link
• Axle track, wheel radius, and corrective gain factors for speeds or ticks (if encoders or wheel speeds are available)

Endorsements


Auve

“Main Street Autonomy is an important component of our plans to scale vehicle deployments worldwide.” [Press Release]
- Andreas Rebane, Board Member and Head of Development


Auve

"The strategic partnership [with MSA] has expedited our technical roadmap by 12 to 16 months.”
[Press Release]
- Ott Männik, Software Lead at Auve


IHMC

“The help that Main Street Autonomy provided was essential for our work. All MSA needed was a little perception data, and they were able to quickly get us the information we needed to have accurate camera data from the robot.” [Press Release]
- Dr. Robert Griffin, Research Scientist at IHMC


CMU Robotics Institute

“Working with MSA for two years has delivered results beyond what we could achieve on our own.”
[Press Release]
- Micah Nye, Chief Engineer and CMU Robotics Institute Masters student

What is a camera calibration?

MSA’s automatic sensor calibration solution works with any number, combination, and layout of perception sensors in any unstructured environment.

check_box Cameras map points in the world into pixels
check_box Calibration tells us how the mapping works
check_box 3D coordinates can be mapped to a pixel location
check_box Pixel coordinates are used to find a ray in 3D space, which can be used to measure the world

Try Calibration Anywhere