CHANNELS

Aerospace

Automotive

Automated & Autonomous Vehicles

Battery & Electrification Technology

Defense

Drone TV

Electronics

Manufacturing & Prototyping

Materials

Medical

Motion Control

RF & Microwave Technology

Robotics & Automation

Sensors & IoT

Test & Measurement

Aerospace
Automotive
Automated & Autonomous Vehicles
Battery & Electrification Technology
Defense
Drone TV
Electronics
Manufacturing & Prototyping
Materials
Medical
Motion Control
RF & Microwave Technology
Robotics & Automation
Sensors & IoT
Test & Measurement
Aerospace & Defense
Search Results

VIPR-GS: Off-Road Autonomous Vehicle Testing

Watch this video to learn more about Clemson's research center, Virtual Prototyping of Autonomy-Enabled Ground Systems (VIPR-GS). Specifically, its autonomous testing facility, the AVL DrivingCube, which allows for the testing of autonomous vehicle driving algorithms in a controlled environment on the chassis dyno.

Topics:
ADAS Automotive Autonomy Unmanned Ground Vehicles (UGVs)

More From SAE Media Group

    Transcript

    00:00:06 I'm I'm the director of the offload autonomy in the Viper GS Center Viper GS Center has a lot of research in V autonomy including offload autonomy spending from sens and perception planning control to even interaction with humans the validation of those auton algorithms are very important people usually use simulation to do validation but when you apply the

    00:00:30 simulation results to the real case it may but may not work the ideal situation should be do new validation and all verification with real vehicles but with real Vehicles experiments are usually very time consuming and costly and some situation may and scenarios may not be possible with real Vehicles due to safety limitations The Driven Cube at clamson here will enable the VAP GS

    00:00:56 Center researchers to conduct the validation and verification of their autom algorisms with real Vehicles so this will help us to push the boundary of the application of the autom algorithms to be closer to the realistic situations hi I'm wenet Croy I'm the Michelin smart State chair professor of vehicle automation at Clemson University the aval driving cube is a fullscale

    00:01:22 test setup that allows us to combine simulated testing and physical testing with the vehicle in the loop uh framework on the sensor enhanced chassis Dynamo so it offers us the ability to take a full scale autonomous vehicle and have it plugged in to the Matrix so to speak so that we can now selectively simulate stimulate and emulate different kinds of

    00:01:48 sensor channels cameras uh Liars Radars or gnss so it creates an effective test stand and we can now create reproducible test environments indoors and so now coupling this with AI based uh workflows for testing we can now uh use these for testing verification and validation of a wide variety of candidate Adas and AD

    00:02:20 algorithms so autonomous driving algorithms um it has proved to be an absolutely uh crucial uh element in the Arsenal of testing and we look forward to um reporting some of the successes uh in the days and years to come my name is B I'm assistant professor at columus University I'm leading a special AI project in the Raa G Center where my team is developing

    00:02:48 realtime AI enhanced weaping for autonomous vehicle to improve situational awareness utilizing AVL driving Cube we will simulate diverse driving conditions and scenarios verifying the robustes of our Visual aggreement and AI sensing capabilities this platform will enable us to validate and refund all AI models offering groundbreaking advancement in Virtual

    00:03:13 design for more robust autonomy hi I'm Seth scof and I am a research associate here at our cuiar campus and Viper GS Center cyber physical test methods play an important role in validating Vehicles equipped with autonomous and assisted driving functionalities the driving Cube combines both real-time simulation and vehicle in theop capabilities on a

    00:03:38 chassis Dyno it aims to speed up the validation and approval process of and autonomous Driving Systems using fully automated scenario based testing increasing test coverage and reducing testing Logistics it's holistic sensor stimulation and full range steering ability while on the dyno provides Proving Ground tasting and brings it to the lab this

    00:04:06 provides safe and replicable testing with coverage of a wide range of driving scenarios and vehicle variants to Clemson's iar campus for autonomous vehicle research the driving Cube includes a vehicle camera stimulator for realtime scenario visualization and a dynamic steering for Simulator for on dyo vehicle steering additional sensor Suite

    00:04:29 stimulation such as gnss for satellite navigation radar for object detection lar for Point Cloud injection and v x for Internet of Things interconnected devices and the environments are all easily upgradeable capabilities for the AVL driving [Music] kit