The Robotics Institute
 Search the site
RI | People | Todd Jochem

Text only version of this site

Todd Jochem
Systems Scientist

No longer a member of RI.

Email address: tjochem@foster-miller.com

Jump to: Research interests | Keywords | Labs & Groups | Projects | Publications

Research interests

The Goal: My goal is to develop deployable systems which allow cars, trucks, and busses to autonomously operate on public highways. To accomplish this, many individual components must smoothly work together.

The Demo: We recently showcased our efforts in this area at the National Automated Highway System Consortium Demonstration in San Diego, CA. During the demo, 3 passenger vehicles and 2 transit busses (Navlabs 6-10) we outfitted with our driving system demonstrated that vehicle based technology was a viable solution for improving our nation's highway system. The system we developed demonstrated lane keeping, lane departure warning, lane changing, headway maintenance, obstacle detection and avoidance, and vehicle to vehicle communications functions.

The Research: For this system, my research involved integrating obstacle and road position information for robust headway maintenance and obstacle detection. The subsystem that was developed provided coverage around all sides of the vehicle and used 6 sensors - a forward radar, 4 side looking radar, and rear laser scanner. Using this subsystem, our cars and busses were able to maintain a safe headway, detect upcoming obstacles and either swerve around them or stop, and sense approaching cars from the rear and change lanes to move out of the way.

The Application: Now we need to move this technology from the test track to vehicles for the average consumer. On this front I'm very interested in operational tests using components of the driving technology like lane departure warning and headway maintenance. It is my goal to have parts of our technology on consumer vehicles by the year 2000.

The Future: In the very long term, I'm interested in using vision based focus of attention techniques to allow autonomous vehicle to execute city driving tasks like intersection navigation.

Research interest keywords

active perception, computer vision, mobile robots, obstacle avoidance, and sensor fusion

Past Labs & Groups

NavLab - Autonomous Vehicles and Driver Assistance
 

Past Projects

AUtomotive Run-Off-Road Avoidance system - To help avoid highway accidents, we have developed a vision-based lateral position estimation system called AURORA.
Autonomous Land Vehicle In a Neural Network - Neural network navigation
Massively Parallel Road Following - Autonomous roadway navigation using video images
No Hands Across America! - Vehicle drives across the United States
Rapidly Adapting Lateral Position Handler - adaptive steering
Run-Off-Road - The Run Off Road Collision Countermeasures program assists drivers by monitoring the vehicle's position in the lane while a person drives.

Recent publications [View all 26 publications]

The Robotics Institute is part of the School of Computer Science, Carnegie Mellon University.
For updates and comments, please see these instructions.
This page maintained by robotwebmaster@ri.cmu.edu