Developing Advanced Collision Warning Countermeasures
printed in the Summer Issue of the Intelligent Vehicle Quarterly)
By now, Ron Hochnadel is a familiar presence at conferences and committee meetings addressing active safety systems. He has discussed his concept for enhanced collision countermeasures using a Look Ahead Detection (LAD) system with many researchers and automotive product developers, and the concept was presented at the recent ITS America annual meeting. Hochnadel's company, INTRASS, was formed several years ago to explore LAD systems; he now sees sufficient promise such that he has relinquished his "day job" to pursue this initiative full time.
As Hochnadel puts it, INTRASS is in the business of rear-end accident collision warning, with the LAD system improving traffic safety by reducing the number and severity of rear-end collisions, including the severity of 'mega' accidents due to low visibility in fog, snow and storms due to sand.
LAD is simple in concept -- it acts as a longitudinal collision avoidance system created by forming a linear network of vehicles in a lane and passing deceleration information back along this network. The information is processed in each vehicle and where warranted, warnings are made. In today's traffic, even with a collision warning system, if a vehicle three cars ahead brakes hard to avoid hitting that stray dog, you (and your warning device) won't know about it until the vehicle directly ahead of you brakes, possibly creating an unavoidable accident -- the LAD concept would provide information on that first braking incident directly to the collision countermeasures system in your vehicle, providing more response time.
Present radar based headway detection systems are designed to eliminate approximately 50% of rear end collisions. Hochnadel
refers to these systems as "first-generation collision avoidance systems." "LAD systems are designed to complement these headway detection systems by looking farther ahead in traffic," he says. "My goal is to eliminate an additional 10% to 20% of all rear end collisions."
The INTRASS LAD system is designed to be effective for all types of vehicles on all types of roadways. As the LAD technology provides information about vehicles ahead of the immediately leading vehicle, the accumulated human reaction time is eliminated in the accident equation. Hochnadel says this is an innovation which is capable of eliminating rear-end collisions which today cannot be prevented by any existing technology or combination of available technologies. "The key is the timing of availability of braking information from the front of the line," he notes. "With LAD technology this information is available almost 'instantly'. Without LAD technology it is available only after a delay of about a half of a second for each vehicle through which it must pass."
Hochnadel readily notes that application of the product requires universal implementation -- the components must be present on many vehicles, ideally all vehicles on the road. He expects that regulation is probably required. A key question is what 'penetration' is required for payback or in other words, what 'penetration' into the vehicle population is required to produce measurable benefits? Hochnadel is studying this issue and plans to publish a paper with his results.
How to get there from here? The best implementation option for LAD technology, according to Hochnadel, is integration with existing Headway Detection and Cruise Control systems, followed by evolution of these systems into a combined LAD/Adaptive Intelligent Cruise Control system with industry standards. "My plan is that Intrass will be the keeper of LAD standards and that the technology will be licensed to the many vendors of Headway Detection and AICC technology as well as to the providers of these systems to OEM vehicle manufacturing companies," he says.
From a technical perspective, some of the primary issues include the communications between mobile units, the hardware and software for the units, and the displays. Vehicle to vehicle network issues in particular need to be researched, particularly the challenge of identification of the nearest front neighbor. Hochnadel envisions that all LAD units will share a single frequency, possibly with other services, with signal collisions being resolved by an Ethernet-type of collision rebroadcast concept. In this scheme, if there is a packet collision, it is OK; the information will be discarded since there will be another transmission almost immediately. He also expects that some existing transponder technology may be utilized.
In operation on the road, as soon as a nearest front neighbor is identified, the concept calls for a unique node identifier to be created, with a model of the network and surrounding environment built and maintained by each vehicle.
INTRASS proposes a 'stoplight' display for the human interface. This familiar format makes the concept easy for the driving public to understand and accept. A green lamp would indicate a linking to the electronic chain and subsequent reception of a data signal. A yellow lamp would indicate a strong braking action in the chain occurring somewhere ahead of a followed vehicle and a red lamp would indicate braking by the immediately followed vehicle. Only one lamp would be illuminated at a time.
Hochnadel expects that LAD systems will most likely be part of the second phase of the Intelligent Vehicle Initiative and will show up in a year or so in tests on selected toll or limited access roads, where the vehicle population can be controlled so 'penetration' can be maximized. During the summer, Hochnadel will be working to assemble a LAD consortium, which he hopes to have in place prior to the ITS World Congress in Toronto this November.
(For more information, contact Ron Hochnadel at email@example.com
or visit the INTRASS web site at www.intrass.com).