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

Intelligent Co-Pilot

A new semiautonomous safety system developed by Sterling Anderson, a Ph.D student in MIT's Department of Mechanical Engineering, and Karl Iagnemma, a principal research scientist in MIT's Robotic Mobility Group, uses an onboard camera and laser rangefinder to identify hazards in a vehicle's environment. The team devised an algorithm to analyze the data and identify safe zones - avoiding, for example, barrels in a field, or other cars on a roadway. The system allows a driver to control the vehicle, only taking the wheel when the driver is about to exit a safe zone.

Topics:
Automotive Cameras Communications Imaging Lasers & Laser Systems Monitoring Motion Control Photonics Robotics Sensors Transportation Visualization Software

More From SAE Media Group

    Transcript

    00:00:05 Humans make mistakes. When humans control machines, the rate and severity of those mistakes increase. In 2010, according to the National Highway, Traffic and Safety Administration there were almost 5 and a half million traffic crashes, in which over 32,000 people were killed and more than 2 million injured. Most of these accidents were caused by human error. When you think about vehicle safety, you might think of seat belts, or airbags, or maybe even anti-lock brakes. This has been the traditional context for the term. In recent years, the emphasis of vehicle safety has shifted

    00:00:39 from collision mitigation, reducing the harm done when accidents happen, to collision avoidance, preventing those accidents from ever happening in the first place. I'm working with MIT's Robotic Mobility Group on a semi-autonomous driver assistance system, that effectively acts as an intelligent co-pilot: when you are driving safely, the system runs in the background, monitoring the vehicle's environment and your performance; but if you make a mistake, one serious enough to cause a collision or loss of control, it intervenes to ensure you avoid it. We've tested our system, in controlled environments like this field, over 1200 times, with over 30 different drivers and very few collisions.

    00:01:17 We're moving towards something that could end up in your next car. The system uses an array of sensors, including an onboard camera and a laser rangefinder, to monitor the vehicle's environment, looking for potential threats, and its algorithms combine all this data into a single metric, identifying a safe field within which the vehicle can operate —essentially a corridor of safety that the car should remain within. As long as the car stays in this corridor, the automated system stays quiet. But if, as you can see here, the driver tries to perform a dangerous maneuver that would cause the vehicle to lose control (as shown by the gray vehicle) the system adjusts the driver's steering commands

    00:01:53 (to the degree indicated by the green bar on the right) to keep it in the safe corridor. The blue vehicle shows the performance of the same driver with our system running in the background. The system is similar to, but different in one key way, from the autopilot systems used by many commercial aircraft. As you may be aware, these autopilot systems are programmed to control the airplane autonomously unless something goes wrong. In the event of a malfunction or unexpected problem, the system shuts down and transfers control to the human pilot. This abrupt transfer of control, particularly in fast-paced or high stress scenarios, where humans tend to perform poorly, can be tricky.

    00:02:29 Many aircraft crashes in recent memory have been caused by human error in times like these. Our system is designed to take the opposite approach: rather than control autonomously by default, the driver remains in control, and the autonomous system only takes over as necessary, and when absolutely necessary to avoid collisions or losses of control. This allows us to exploit the automated system's ability to respond quickly and precisely to well-defined control objectives, while exploiting the driver's uniquely human ability to detect and contextualize patterns and new information, reason inductively,

    00:03:03 and adapt to new control modes, as necessary while driving.