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

Obstacle-Avoidance Algorithm for Autonomous Vehicles

Stanford University engineering students have been testing an obstacle-avoidance algorithm, using a pop-up obstacle they rigged up from a tablecloth and a leaf blower. They have been using X1, a dune-buggy style car that students built for testing autonomous driving programs, to develop the algorithm and see how well it could avoid a pre-programmed obstacle. The students recently tested the algorithm and obstacle at a raceway several hours north of campus. The X1 algorithms swerved the car perfectly to avoid the obstacle while driving right at the edge of the car's handling abilities. The next step involves porting the software into Stanford's autonomous Audi TTS race car and other vehicles, to test obstacle avoidance on the track and under more complex environments and pressures.

Topics:
Automation Automotive Motion Control Robotics Software Test & Measurement Transportation

More From SAE Media Group

    Transcript

    00:00:00 Stanford University. Today we're here at Thunderhill. It's a great place to test a lot of things that you wouldn't be able to do on a normal public road. We're really exploring autonomous vehicle control, right up to the limits of what these cars are capable of doing. We're using x1, which is a completely student built

    00:00:26 vehicle test bed platform. And we are running it autonomously around an oval track on the skid pad. At the edge of one of the tight turns, the vehicle is operating at the edge of its vehicle handling limits. And then an obstacle will pop up, and the vehicle has to adjust its trajectory. What we built was our own sort of airbag, by using plastic tablecloth and sewing it into a tube.

    00:00:55 And we use a leaf blower and place the tube across the entire lane, making the car be forced to make a lane change. What we're really trying to test here is a control algorithm that can balance a lot of very challenging things that crop up when a car has to approach its physical capabilities. This controller is trying to not hit the obstacle, but it's also trying not to spin out. Of lesser priority is trying to do this smoothly,

    00:01:24 so that and occupant isn't jerked around. And so there's a lot of conflicting objectives that have to be met. The car is making all these decisions itself. It goes into a turn and it's trying to use all the available grip. If it starts sliding, it can recognize that it's sliding and try and recover from that. And the hope is that if we can figure these things out in a controlled research environment like this,

    00:01:45 we can extend those ideas to production vehicles that may have autonomous capabilities in the future. If a car is physically capable of avoiding something, we'd like to develop the controllers that make that happen. For more, please visit us at stanford.edu.