Reduce the number and severity of auto accidents, using current technology on human-driven cars
We have produced vehicle to vehicle technology and an associated patent portfolio that will enable vehicle coordination with local information based on easily achievable sensors that are part of the vehicles themselves.
The system can be much cheaper to deploy and operate than designs of record from transportation authorities, by bypassing dependence on network carriers and terrestrial location sensing. The system further assumes the need for location sensing and reaction that are hundreds of times more accurate and faster than plans of record technologies can achieve. None of the techniques covered in our technologies require central infrastructure investment or all vehicles to participate to succeed.
AUTONOMOUS VEHICLES
Will self-driving cars eliminate accidents? Maybe, but not for 30 years. They will cost considerably more than today’s cars. Many drivers do not want to become merely passengers in their own cars. Today, that technology is far from working, despite billions of dollars spent.
WHAT IS V2V?
The concept behind “V2V,” or vehicle-to-vehicle communication, is simple. Cars communicate key information, such as their own location, heading and speed, on a peer-to-peer basis, 10 times per second. Other cars listen to this information and then use it to prevent or mitigate accidents.
Beyond this conceptual simplicity, the problem is quite complex. That is why we created this website, which explains both the technical problems and their solutions.
How is accurate location determined, to
4-inch accuracy?
There have been many prior approaches to determining accurate vehicle location.
GPS. Most people assume that GPS will provide location. However, GPS is accurate to only 30 feet (on a good day). A location accuracy of 4 inches is required to prevent accidents.
Differential GPS. While “differential GPS” is highly accurate, it works only with a local fixed transmitter, often called Road Side Equipment, that continuously transmits short-range “corrections” to received GPS signals. Even if tens of millions of these boxes were to be installed, they still won’t work because Differential GPS is not compatible with moving vehicles.
In addition, GPS cannot solve the problem of overpasses and underpasses because elevation information in GPS is accurate to only about 100 feet.
Comparing video images from the car with stored images. This approach has been tried for years and cannot be made to work. Wind on foliage, clouds, sun angle, construction, and crop growth all make “comparing images” unreliable. Also, the amount of storage needed, frequent image updates, and non-real-time data make this approach infeasible. Ever driven past farm land? It all looks the same to you, and also looks the same to video comparison algorithms.
The truth is that the autonomous vehicle people have been working, without success, on this problem for years. This problem is the number one reason you can’t buy a self-driving car today. The car cannot reliably or accurately know where it is.
HOW DO YOU SOLVE THIS PROBLEM?
We take an entirely different approach. Others try to get an absolute location on the surface of the earth. However, to keep two cars from hitting each other, we only need to know the relative location of nearby vehicles. We compute a relative location between all nearby vehicles.. Cars “negotiate” 10 times per second so that all nearby cars are in full agreement on all of their respective location. We still use GPS, but only for course location, not for fine location.
How to avoid hacking?
People are rightly scared that any radio-based safety system could be hacked, either making the system inoperative or worse, cause accidents to happen. Others have proposed that all communications be encrypted. There are numerous reasons why encrypting all communications will not work. How, and who, will distribute hundreds of millions of digital keys? How do you keep bad people from using good keys? In addition, encryption makes messages too long to transmit and receive quickly.
HOW DO YOU SOLVE THIS PROBLEM?
We use an entirely different approach. First, data is not encrypted (with a few exceptions). This “clear text” Is the most reliable way to avoid hacking because there is nothing to crack, as any cryptologist will tell you. All communications are both local (about half a mile) and time-limited (about a second). A hacker would need to be close by and would have to modify his bad data 10 times per second. Even so, a set of other checks still block hacking. In fact, any hacking attempts are immediately caught in real time (by every car) and forwarded to both other nearby cars and law enforcement, including pictures and the precise location of the miscreant. Since all communication is broadcast in clear-text, everybody hears everything. Hacking attempts are caught, flagged, and blocked in less than a second. Attempted hackers have no reward, except jail time.
In any case, vehicles always use data from their own internal sensors as priority over any received radio data.
How do you preserve privacy?
All other approaches use some kind of permanent radio or vehicle ID, such as an IP address, VIN number, phone number, or social security number. All of these, by their very nature, and are neither private nor confidential.
HOW DO YOU SOLVE THIS PROBLEM?
We use no permanent ID of any kind. Vehicle ID is solely the location of vehicles, which is updated 10 times per second. Since no two vehicles can be in the same place at the same time, this method of vehicle ID is always unique, at any moment in time. This provides at least the same level of privacy as standing on a sidewalk observing traffic as it drives by.
The technical word for privacy is anonymity. Because we use no permanent ID, the safety features of the system are available without disclosing who you are.
We do use any centralized authority. There is no registration, no number assignment, and no fees.
How can 100+ cars all communicate at the same time?
There is currently dedicated V2V communication bandwidth assigned by the FCC. In addition, there are unregulated “Industrial, Scientific and Medical” bands, such as used by Wi-Fi. There is also an unregulated band just for government use. We assume, for our calculations, that only the currently assigned FCC band is available. Most of our technology will work in any radio band, however. The type of digital radio we use is well-proven and similar to existing Wi-Fi radios (but do not use Wi-Fi communication protocols).
It is widely accepted that 180 vehicles need to communicate simultaneously. As examples, consider two freeways crossing or an airport parking lot. Other proposed V2V protocols, such as IEEE 802.11p, cannot provide nearly enough capability. These proposals start to fail at about seven vehicles and fail completely with forty vehicles.
We can support 600 vehicles simultaneously, reliably, in the currently assigned bandwidth using existing digital radio technology.
HOW DO YOU DO THAT?
We use a number of methods. Primarily, we do not send unnecessary information, such as IP addresses, cryptographic keys, or commercial advertisements. We use application-specific data compression to keep message size to an absolute minimum. We eliminate unnecessary dead-time between messages. Because of our core reliability and no wireless “collisions”, we do not have to resend messages, like Wi-Fi does.
WHAT ABOUT EMERGENCY VEHICLES?
Emergency vehicles have guaranteed priority. Normally, there is plenty of bandwidth for everyone. However, in major emergencies such as floods and wildfires, there may be over 100 emergency vehicles responding. If necessary, emergency communications can take up to 100% of bandwidth. There are no “dedicated” emergency bands, as in other proposals. Such fixed spectrum allocation both fails to provide sufficient bandwidth for emergencies and also locks out perfectly good bandwidth for regular people, the other 99% of the time.
How are detailed maps made and distributed?
Traditional maps are static. They do not include construction, stopped UPS trucks, or potholes. Even if there were some massive map distribution system, there are still three problems: (a) such maps will never be real-time, on the order of seconds. When you need a map the most, it will be out of date, maybe by weeks. In addition, there is no possible way to get enough bandwidth to send and receive such maps in real-time. (b) Who will create and who will pay for these maps, even if they could be created and distributed? (c) Current maps are good only to GPS accuracy, about 30 feet. For preventing collisions, a resolution of 4 inches is necessary. Such maps do not exist today. If they did, they would need thousands of times more memory than maps do today.
HOW DO YOU SOLVE THIS PROBLEM?
We solve the “map problem” using one of our more creative technologies. Every vehicle builds a map in real-time as it travels. Such maps contain only local information needed for V2V safety. There are no ads for nearby hamburgers. These maps are shared, constantly, in real-time, on a peer-to-peer basis with other vehicles, as they pass each other. Only unused bandwidth is used for map sharing, so there is no loss of bandwidth for more critical data, such as an impending collision. A novel “bidding system” means that only the most recent, local and accurate map portions are shared, dramatically reducing map distribution bandwidth. (All vehicles in range pick up new broadcast map data automatically.) Maps are not absolutely necessary for our system to work and prevent accidents, but they are helpful for many other purposes, such as identifying construction, delivery vehicles, and sidewalks.
Thus, every vehicle is constantly creating and updating maps 10 times per second. Every vehicle shares its map with every other vehicle, on a request basis. (There is no “push” of unneeded map updates.) In addition, our maps are based entirely on “where vehicles actually drive,” not where lines are painted (or not painted) on the roadbed.
How does the system benefit a few early adopters?
Other proposals have a seriously bad “chicken and egg” problem. Those proposals would provide no realistic collision protection until at least 90% of all vehicles are equipped. This will take 10-15 years, AFTER the government mandates that all new vehicles be equipped, which currently the Federal government (DOT, FCC, NHTSA) has no intention of doing. This means that most new car buyers would be asked to pay for something of no value to them over their ownership period. This is a total non-starter for the public.
HOW DO YOU SOLVE THIS PROBLEM?
We provide several immediate benefits to early adopters.
For example, all vehicles detect and broadcast the status of nearby parking spaces. If there is an open parking space within half a mile, you will know about it. If a parking lot or street parking is full, you will know that, too. When 1% to 3% of cars are equipped, the status of nearby parking spaces will be known to other early adopters.
As a second benefit, all vehicles detect and broadcast road status information 10 times per second. These broadcasts have a range of about half a mile. If there is a dog in the road, stopped traffic ahead, or an emergency vehicle approaching, you will know. Emergency messages are “forwarded” so they reach additional drivers. Even with less than 1% of vehicles equipped, emergency information will typically be available to all equipped vehicles.
A third benefit is what we call “courtesy messages.” For example, one driver may inform another driver that his brake lights are out. These messages are lower priority so they never block real-time safety information. These messages are not social networking, chat, web pages or advertisements. The entire V2V system is “closed,” meaning it does not send or receive messages from non-V2V sources. Even with 1% of vehicles equipped, you are likely to receive such courtesy messages.
A fourth early adopter benefit is the use of fixed location, government-managed transmitters. This is a special type of V2V device that uses cryptographically signed messages to assure validity. These may be used in emergencies, large construction projects, detours and the like. Every equipped vehicle receives these “warning beacons” when within range, even if it is the only equipped vehicle on the road.
How can the system be super-reliable, fast, and predictable?
Latency is the time between when data is known by a sender and when it is actually received by another vehicle. Network protocols like W-Fi (and IEEE 802.11p) have all vehicles transmitting at once. If two messages overlap each other in time (a “radio collision”) they are both thrown away, and each transmitter tries again at some later random time (CSMA). Such networks have inherently unpredictable latency, throughput, and performance. We use a technology known as “time-slots” (TDMA). Every vehicle in range is assigned (by consensus of peers) a fixed (for a few minutes, typically) time-slot. No other vehicle can use your time-slot, so there is no corrupting your messages by another transmitter. In addition, there is no “third party,” such as a base station, road-side equipment, or a cellular tower, to receive, store, delay, and possibly corrupt your messages. Thus, the only delay between a transmitter and receiver is the speed of light between the two vehicles. Only a true peer-to-peer network can do this.
Our protocols include dynamically managed transmit power to reach the maximum range (number of vehicles) while maintaining sufficient power for highly reliable reception.
Vehicles do not send ACK and NACK messages. All messages are broadcast in clear-text to all other vehicles in range. There is no “set-up” and “tear-down” time, as in other protocols, including Wi-Fi, IEEE, and cellular protocols. In addition, there no ID recognition and validation delays for billing purposes.
All aspects of our technology were purpose-built for vehicle safety. Which is unlike Wi-Fi, which was designed to carry a few weather updates per day between Hawaiian Islands, and now to load email. It is also unlike cellular, which designed to carry voice calls tower-to-phone (e.g., point-to-point, not broadcast), at least when your network provider is not too busy.
Calculations involving latency, reliability and predictable performance are quite technical. Contact the authors for such mind-numbing detail.
Will there be monthly fees?
“5G” is not a technical specification. Rather, it is a marketing term to get people to buy new hardware.
NO. The cellular companies want you believe that “5G” will solve every problem there is. Additionally, they expect that every vehicle will pay $130 per month, forever. Otherwise, you get nothing. Your car might not even start. The government (NHTSA, US DOT) agrees with this approach. Nobody else agrees with the idea of mandatory fees, particularly consumers.
OUR TECHNOLOGY IS PEER-TO-PEER. No subscriptions or even registration is needed. No cellular company or government agency is involved. No infrastructure needs to be built and no road-side equipment needs to be installed. No changes to existing law are necessary. It works everywhere, right out of the box. All we need are people to build the boxes.
Can I retrofit my existing car?
YES. Our technology is low-enough cost that retrofits are reasonable and likely. You won’t have to buy a new car to get safer driving. New cars may have additional capabilities that a retrofit won’t necessarily have.
Doesn’t this exist already?
No, although you may have seen adverting implying that it does.
Cellular companies will never be able to deliver V2V safety communication. Latency is too high. There is not now and likely will never be enough cellular coverage for a safety system. And, high, required monthly fees are a non-starter for just about everyone. In addition, cellular communication technology is directed only to the digital radios, not to all the other key technologies necessary, like hacking prevention, privacy, accurate location, and usable maps.
There are “standards,” such as IEEE 801.11p. There are two problems with these standards. First, they don’t work. For example, IEEE 802.11p maxes out at about 40 vehicles. Second, these standards only address about 5% of what is needed. They describe only the digital radios, which have been around for years. Like the idea of using cellular, they fail to address, much less solve, the most important problems of V2V safety communication.
Will this work for non-cars, like bicycles and drones?
Yes. The technology is small enough that it can be used for pedestrians, bicycles, skate boarders, sheep, pets, watercraft, aircraft and drones. It is not enough for cars to not hit each other. We don’t want cars to hit pedestrians either, even if they are staring down at their phones.
Won’t self-driving cars solve this problem?
Good question. Many people think that autonomous vehicles will prevent accidents, once we are all driving them. By the time everybody is riding as a passenger in self-driving car, we should indeed see fewer accidents. This is not likely to happen in your lifetime. Currently, self-driving cars are completely “autonomous,” meaning they accept no input from outside the car. (Although they will likely use our technology in the future.) What they can’t see, they can’t avoid. For example, they cannot see the car at a blind intersection about to run a red-light and T-bone you. They cannot see a crazy driver crossing the double yellow line on a mountain road, about to hit you head-on. They cannot know about a dog in the road ahead, or ice on a bridge. In fact, for the most dangerous kinds of accidents, they provide little benefit. Also, they won’t tell you about empty parking spaces.
Some vehicles will never be self-driving, such as most construction equipment, bicycles, and water-skiers. However, these kinds of vehicles will benefit significantly from our technology.
Autonomous vehicles for consumers currently don’t exist. When they do exist, they will be expensive. At least half of all drivers say they want to continue to drive their own cars. They just don’t want to be hit by other drivers.
How will this help me if other people don’t have the technology, too?
Another good question. Clearly, avoiding accidents works best if everybody participates. However, there are several ways that having some vehicles equipped helps everybody.
For example, if there is a problem in an intersection ahead, equipped vehicles will know this and slow down. This might annoy drivers behind who can’t see the problem, but they will have to slow, too. A few cars driving more safely help everyone else to drive safer. If a parking lot is full, having fewer people driving around and around makes it faster for others to get in, fail, and get out. It means fewer people driving around the block, too.
When some cars slow way down to cross an icy bridge, other drivers will see this and slow, too.
A unique feature of our system is called, “proxying.” This means that a nearby unequipped vehicle will be “proxied” by the closest equipped vehicle. The equipped vehicle broadcasts V2V safety data for the unequipped vehicle as if it were equipped. This multiples the effective percentage of equipped vehicles. For example, if one equipped vehicle is proxying four others (in front, behind and the sides), this multiplies the effective percentage of equipped vehicles by five. When about 20% of all vehicles are equipped, the effective coverage includes most vehicles in traffic.
In a novel approach, compared with both other proposed technologies and autonomous vehicles, vehicles will warn other vehicles of unsafe conditions, even when it might not affect them directly. For example, if a car is about to run a red-light, all equipped vehicles that can see that car will issue an immediate warning. All equipped vehicles will then know not to enter the intersection. This prevents other vehicles from entering the intersection, too, even if they are not equipped. An equipped vehicle will automatically sound its horn, when appropriate. As car horns were designed to do when invented 112 years ago, this warns all drivers within earshot of the horn they had better show immediate caution.
Why won’t 5G be the solution?
5G is a marketing term. It does not actually exist, yet, because the specification for the Standard is not done. There is a “low-speed” 5G that does work and has been implemented. This “low-speed 5G” is actually slower than current 4G. Even when real 5G is finally deployed, it cannot provide V2V safety for at least the following reasons.
Monthly fees are projected at $130/month per vehicle.
Use of phone numbers means there is no anonymity.
Central authority means that the cellular companies own the system, not you.
Use of cell towers means all communication vehicle-to-vehicle is delayed. This delay latency is too high.
Use of cell towers (base stations) means all communication is point-to-point, not broadcast. For 100+ vehicles near one tower, there will not be enough bandwidth.
Cell tower coverage includes many urban areas, but a small fraction of all roads. The most dangerous roads often have no cellular access. Who wants a safety system that works some of the time?
5G does not provide most of the necessary features, such as accurate location, lane maps, broadcasts, or security. It is essentially just a digital radio.
What is the technology and how much does it cost?
The technology was invented by some guys at Zetta Research, LLC. So far 19 patents have issued for an IP suite directed to a single goal: eliminating and reducing vehicle accidents using peer-to-peer vehicle-to-vehicle (V2V) communication.
This technology is not suitable for advertising or social networks. It is not related to cellular communication. It is purely for vehicle safety.
Performance has been extensively tested via simulation.
We are proposing license fees of one dollar per car. That is one time – not monthly.
The underlying technologies, such a digital radios, cameras, and computers, are well-proven. The novelty resides in details of communication protocols and algorithms.
We do not sell directly to consumers. Rather we license this technology to others, such as automotive companies, transportation companies, and chip companies. We have already licensed to a multi-national Asian company.
We are not patent trolls – waiting around for someone else to invent the same thing and then suing them. No, what we want is to prevent the 18,000 accidents PER DAY in the US, 1.25 million deaths per year, 10 millions serious injuries, and $300 billion per year in direct out-of-pocket costs per year, in the US alone.
The only people who are against this are the cellular companies, who think that can stick owners for $130 per car per month, forever. And the car insurance companies, who do not want to be put out of business. When there are no accidents, there will be no need for auto insurance.
Technology comparison table
