About Iris
IRIS.Pathfinder

IRIS.PATHFINDER will plan and execute collision-free motion on most robots and hydraulic machines. Set a destination and the manipulator will get there in minimum time.

This tool is great for adding real-time supervision to open-source code or AIs, which might not consider all the physical limitations, resulting in costly mistakes.

01

COLLISION AVOIDANCE
Connect a CAD model or 3D sensor representing the environment, manipulator, or payload and IRIS.PATHFINDER will continuously check for collision and plan around obstacles. It will do this even if experimental code, a notice operator, or an AI command a collision.

02

SINGULARITY AVOIDANCE
A common problem many of us have faced is two of your robot’s joints are lined up and movement grinds to a crawl while the joints exchange angles. Our controller handles this situation in stride without noticeable slowdown, smoothly passing through a singularity.

03

MODEL-PREDICTIVE CONTROL
Swinging a heavy payload at maximum speed strains all joints in the manipulator as forces pass through them. What might be suitable for the largest joint could overwhelm the smaller joints. However, simply slowing down to accommodate will cost cycle time and doesn’t always work. IRIS.PATHFINDER considers these factors in real-time and re-works trajectories to ensure all parts of a manipulator stay within their performance envelope. Especially for large industrial robots or hydraulic manipulators, this means that sharp corners, which otherwise can cause problems, are adjusted to the shortest time solution that keeps the motion within specification.

IRIS.XR

In industrial settings where human decision making is still superior to autonomous machines, tradeoffs are no longer needed. It lets an operator take over to perform the challenging parts or intervene if the AI or algorithm gets stuck. Our unique control technology connects the user’s senses and actions directly to the robot using a VR and motion/haptic control system, making the machine “feel” like a natural extension of the operator’s body.

Our goal with IRIS.XR is to simplify how the operator conveys intent to complicated machines, reducing cognitive load. This makes controlling of any robot or machine easy and intuitive, requiring almost no training.

01

CO-AI
Collaborative AIs are customized for your application and share the body of the robot with the operator, allowing each to focus on what they’re best at. A co-AI can be used to minimize cognitive load for the operator by inferring intent or it can autonomously handle the repetitive work, freeing the operator for other tasks.

02

EXTENDED REALITY
IRIS.XR combines several emerging concepts to put the operator into the machine’s world. It combines the interface hardware of virtual reality with the UI/UX elements of augmented reality. Our proprietary software between the robot and operator allows for real time teleoperation over the internet.

03

MINIMAL TRAINING
With a focus on capturing and translating operator intent to machines, IRIS.XR is very easy to learn. Most novice operators achieve control proficiency in minutes. In a trial of 43 novice operators, we saw a 20% improvement in cycle times after less than a minute of practice.

COLLISION AVOIDANCE

Connect a CAD model or 3D sensor representing the environment, manipulator, or payload and IRIS.PATHFINDER will continuously check for collision and plan around obstacles. It will do this even if experimental code, a notice operator, or an AI command a collision.

 

SINGULARITY AVOIDANCE

A common problem many of us have faced is two of your robot's joints are lined up and movement grinds to a crawl while the joints exchange angles. Our controller handles this situation in stride without noticeable slowdown, smoothly passing through a singularity.

MODEL-PREDICTIVE CONTROL

Swinging a heavy payload at maximum speed strains all joints in the manipulator as forces pass through them. What might be suitable for the largest joint could overwhelm the smaller joints. However, simply slowing down to accommodate will cost cycle time and doesn't always work. IRIS.PATHFINDER considers these factors in real-time and re-works trajectories to ensure all parts of a manipulator stay within their performance envelope. Especially for large industrial robots or hydraulic manipulators, this means that sharp corners, which otherwise can cause problems, are adjusted to the shortest time solution that keeps the motion within specification.

Contact Us
Contact Us

CO-AI

Collaborative AIs are customized for your application and share the body of the robot with the operator, allowing each to focus on what they’re best at. A co-AI can be used to minimize cognitive load for the operator by inferring intent or it can autonomously handle the repetitive work, freeing the operator for other tasks.

 

EXTENDED REALITY

IRIS.XR combines several emerging concepts to put the operator into the machine’s world. It combines the interface hardware of virtual reality with the UI/UX elements of augmented reality. Our proprietary software between the robot and operator allows for real time teleoperation over the internet.

 

MINIMAL TRAINING

With a focus on capturing and translating operator intent to machines, IRIS.XR is very easy to learn. Most novice operators achieve control proficiency in minutes. In a trial of 43 novice operators, we saw a 20% improvement in cycle times after less than a minute of practice.