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

Student-Designed Bots Mimic Self-Driving Cars, Navigate Miniature City & 'Rescue' Animals

Stanford University  students programmed robots to autonomously navigate an unknown city-scape and aid in a simulated rescue of animals in danger, in a robotics class that mimics the programming needed for autonomous cars or robots of the future. The small robots were outfitted with laser sensors to help them locate and record obstacles, cameras to enable detection of pictures of animals, and an on-board computer. The student teams programmed their robots to work at varying levels of autonomy, using industry-standard software and image classification developed through deep learning algorithms.

Topics:
Automotive Imaging Motion Control Sensors

More From SAE Media Group

    Transcript

    00:00:00 [MUSIC PLAYING] This class is about autonomous robotics. The competition today, we've got several teams with different Turtebots. And so, we've been tasked to explore the neighborhood and look for animals that might be in need of rescue. These robots are small, but they comprise most of the sensors that actually you would see on real self-driving car. One thing that I'll have to do is

    00:00:32 use a LIDAR, which is this kind of a spinning laser on top to map its environment. The robot will figure out a way to navigate throughout the map without hitting anything. At the same time while its building up the map, it's also localizing the animals. So it's using the cameras and machine learning to determine what objects are what, figure out which ones are animals and what are trees and people.

    00:00:56 We made sure that the base requirement of the project were kind of small and that everyone could tackle them but then give time to students to develop their own extensions. So some teams are able to react intelligently to a bicycle crossing the road. Some other teams will be extremely efficient in the way that they'll choose which animal to go rescue. Of course, we're not dealing with all the complexity of a real self-driving car, but the point to discuss

    00:01:22 is to make that students know all the key technical aspects. It's definitely given me a healthy dose of respect of what real autonomous cars have to do. Here, it's funny when we mess up and misclassify something, but in a real world situation, there can't be any of these mistakes or any of these miscalculations. I knew a little bit about everything before but being able to implement it on a real robot and integrating all of these ideas,

    00:01:51 putting them all together has been really fun. [APPLAUSE] [MUSIC PLAYING] For more, please visit us at Stanford.edu.