Streamlining the connection to trained ML models with the ONNX Helper Library
There are many cases where it’s desirable to incorporate trained machine learning (ML) models into your simulation model. The following are some concrete examples, based on our
ML Testbed cases:
-
Replacing static or distribution-based travel times with an ML model, trained on real-world data, that uses date and time inputs to predict travel duration.
-
Incorporation of the same ML model used in a real-world refurbishing facility, used to classify the repairability of arriving components, into a simulation model of the facility for increased accuracy.
-
Showing visually and statistically the impact and overall performance of implementing an ML model trained to control machine speeds (e.g., using reinforcement learning) prior to it being deployed in the real world.
We thought you would be interested in learning that this system now has its digital twin!
We have just released three new virtual experiments from the rapidly growing Quanser Interative Labs collection -
Using QLabs, students follow the same process as if working with the physical system in the lab:
-
development of the system's mathematical model
-
implementation and simulation of the system's dynamic behavior
-
controller design
-
controller validation on a high-fidelity 3D real-time virtual model of the system
The virtual systems not only look like the real ones, they also behave, can be manipulated, measured, and controlled like real devices. And when students go to the lab, they can deploy their virtually-validated designs on actual physical equipment. Sounds like a digital twinning approach adopted by industry? With QLabs, your students will be ready for it!
Want to have a closer look at QLabs to see how the platform can fit your course or program needs?
The global standard in transformational engineering labs for Controls, Robotics, and Mechatronics, optimized for the academic setting.
Academic institutions trust Quanser to strengthen their reputation through transformative teaching and research labs that accurately replicate theories and bring engineering mathematics and concepts to life.
PIONEERS IN AUTONOMOUS ROBOTICS
The adventure, as Dr. Jacob Apkarian, Quanser Founder and CTO, puts it, started more than two decades ago. From the first stationary systems for research in flight dynamics and aerial vehicles control to the latest aerial and ground vehicles, the open architecture design is what makes these platforms so helpful for researchers. Watch Dr. Apkarian explain why.
Autonomous Vehicles Research Studio Video
Quanser is excited to announce our latest autonomous vehicles solutions
Many unmanned vehicle application areas have emerged that require not only multiple vehicles, but collaboration between multiple vehicle types.
The diversity and complexity of these applications present significant challenges to researchers and developers, such as increasing difficulty in setting up and maintaining the evolving and growing drone fleet, as well as the need for standardized validation and verification platform to facilitate development, evaluation and analysis. Quanser’s new Autonomous Vehicles
Quanser AERO for Essential Dynamics and Control Concepts
The field of aerospace design and control has never been more immediately accessible and exciting as it is today. However, the recreational drones available to students lack the instrumentation and consistent dynamics required to learn the fundamentals of aerospace control and dynamics.
The Quanser Aerospace Solution line offers a full range of mechatronic systems forteaching topics ranging from introductory controls all the way to collaborativeunmanned vehicle systems.
These solutions provide safe, proven, consistent, and accurately modeled systems which provide the foundation for developing the future of aerospace control systems.
The Quanser Aero is a unique platform that allows students to explore and gain insight into the dynamic complexities of flight applications. At its core, the Aero is a high-precision, plug-and-play physical system to study helicopter and quadcopter flight motion and control. Applications include 1 DOF attitude control, conventional 2 DOF helicopter flight, and even quadcopter dynamics and control. Reconfigurable elements let you quickly adapt the Aero to modern applications in mechatronic systems design.
Quanser's Products Offered by Techenware
course- Introduction to Control Systems
Introduction to Control Systems is typically a core course of the engineering undergraduate curriculum in Electrical and Computer Engineering, Mechanical Engineering, Aerospace Engineering, and Chemical Engineering.
Quanser plants offer the right mix of features, precision, robustness, and flexibility for a wide range of course variations and budgets.
With the range of servomotor-based experiment options – from topboards for NI ELVIS platform to QUBE-Servo and other servomotor experiments – you can select what best supports the fundamental concepts taught in your introductory course, including motor characterization and modeling, velocity control, position control, PID control and more.
Most systems come with a comprehensive ready-to-use courseware that allows you to quickly incorporate the plants into your course.
Control of Robotic and Autonomous Systems
Control of robotic systems is a core part of the Robotics and Intelligent Systems courses in Electrical and Computer Engineering, Mechanical Engineering and Mechatronics Engineering.
Quanser plants offer the right mix of features, precision, robustness, and flexibility for a wide range of introductory to advanced courses in robotic systems control.
The systems demonstrate the concepts of robotics, including forward and inverse kinematics, manipulator dynamics, trajectory planning, vision systems and vision guided motion, control of multiple link manipulators and autonomous systems.
Many new research fields emerged recently, applying innovative robotic solutions in areas such as medicine and rehabilitation, defense and aerospace sectors.
Quanser has established strategic partnerships with key industrial robotic companies that allow researchers to benefit from completely open architecture and extremely flexible robotic platforms and enables innovative research to be performed using market leading robotic systems without the limitations of the typical closed controllers
Haptic Systems and Applications
Haptic technology is finding many applications in medical training, rehabilitation, areas requiring handling of dangerous materials, as well as in the entertainment and gaming industries.
Research applications require platforms capable of reproducing movements and forces accurately, yet cost-effectively, through flexible frameworks on open architecture devices.
Quanser haptic devices - offer researchers the ability to easily and quickly change parameters of control systems, swap or customize haptic devices to adapt them to the specific needs of your research team encounter.
Quanser AERO Demo Video Available
The video demonstrating Quanser AERO experimental plant is now available on Quanser website.
The video highlights the main features of the experiment, including QFLEX 2 interfacing panel options, shows different system and experiment configurations, and lists the topics covered by the courseware.
Select the high resolution or low resolution file based on your need and use in your presentations, during customer visits or web meeting.
For more information...
