feature article
Subscribe Now

Robot’s Gambit: Watch Out, the Robots are Coming for Us

This article is for all those people who think that Asimov’s Three Laws of Robotics are real. They’re not, and the robots appear to be coming for us. Two seemingly unrelated articles caught my eye on the morning of July 25. The first article described the unfortunate death of a motorcyclist due to being rear-ended by a Tesla AV (autonomous vehicle) operating in Tesla’s Autopilot mode on Interstate 15 just south of Salt Lake City. The second article discusses how a chess-playing robot in Russia broke a young opponent’s little finger when the 7-year-old boy tried to make a move out of turn.

The first accident occurred after 1 am, so it was quite dark. Motorcycles have only one tail light. They’re often invisible to the eye in the dark, especially if they’re painted entirely in matte black paint. (I don’t know the color of this particular motorcycle.) Obviously the Tesla didn’t see the motorcycle. The Tesla’s driver said he didn’t see the motorcycle either.

This Tesla vehicle accident is the type of incident that’s getting far too familiar. According to a July 18 Reuters report, the US National Highway Traffic Safety Administration (NHTSA) “has opened 37 special investigations of crashes involving Tesla vehicles where advanced driver assistance systems such as Autopilot were suspected of being used.” One of those special investigations involves a crash of a 2021 Tesla Model Y vehicle that killed a motorcyclist on the Riverside Freeway in southern California on July 7, 2021. The report also states that a “total of 18 crash deaths were reported in those Tesla-related investigations, including the most recent fatal California crash.” 

As for the chess-playing robot incident, an article in The Guardian reported on the incident and a video published by Guardian Sport on YouTube shows the industrial robotic arm taking one of the boy’s chess pieces. The boy then makes a quick reply by moving his rook. The robotic arm grabs his finger before he finishes moving his rook and refuses to release the finger. Two adults quickly rush in to free the boy’s hand.

The Moscow Chess Federation rented the robot for an exhibition during the Moscow Chess Open competition. The article in The Guardian quotes Sergey Lazarev, president of the Moscow Chess Federation, who says, “The robot broke the child’s finger. This is of course bad.” The child that the robot injured is ranked in the top thirty players in the Federation’s “under-nine” category. Sergey Smagin, vice-president of the Russian Chess Federation, said “There are certain safety rules and the child, apparently, violated them. When he made his move, he did not realize he first had to wait.” Apparently, in Russia, there are very serious consequences for breaking rules, even in chess competitions. Despite being injured and having his hand put in a cast, the young man returned the next day and finished playing in the tournament.

Although these two incidents may seem unrelated, I think they’re both closely related. In both cases, we’re dealing with robots interacting with people in a mixed environment, with injurious or deadly results. A Tesla automobile operating on Autopilot is a wheeled robot traveling along an Interstate highway at 70 or 80 mph in a field of traffic where most vehicles are being operated by people. One mistake causing two vehicles to occupy the same location at the same time can and does cost people their lives. 

The chess-playing robotic arm is also operating in a mixed environment, although a much smaller one: a chess board. In this case, the robot and the child’s hand occupied the same space at the same time. Fortunately, the young chess player suffered only a fractured little finger, instead of losing the finger. In George Lucas’s original Star Wars movie during another chess-like game, Han Solo told C3P0 that “…droids don’t pull people’s arms out of their sockets when they lose.” Apparently, Han Solo never visited Russia in the Millennium Falcon.

Many readers of EEJournal may be shocked to learn that there are people in our society who believe that Isaac Asimov’s Three Laws of Robotics are, in fact, real, and that robots are benign. Asimov’s three laws are:

  • First Law: A robot may not injure a human being or, through inaction, allow a human being to come to harm.
  • Second Law: A robot must obey orders given it by human beings except where such orders would conflict with the First Law.
  • Third Law: A robot must protect its own existence as long as such protection does not conflict with the First or Second Law.

These three laws are science fiction. They first appeared in Asimov’s 1942 short story “Runaround,” which was published in Astounding Science Fiction, a magazine I still read today that’s now called Analog. Asimov built many stories and novels around his Three Laws of Robotics, eventually positing a “zeroth” law that puts saving humanity above saving an individual. 

Star Trek: The Next Generation is probably responsible for placing Asimov’s three laws into the collective consciousness of an even broader audience because the popular android character named Data had a positronic brain (another Asimov invention, but attributed to Dr Noonian Soong in the TV series) with the three laws embedded in it. Positronic robotic brains and the Three Laws of Robotics are great inventions of fiction, but they’re not real.

However, I do not think either of the above accidents are due to the fictitious nature of the positronic brain or the Three Laws of Robotics. Instead, I think the root cause is a lack of proper sensors. In Tesla’s case, we’re dealing with a low-visibility situation. A motorcycle rider in the dark is rather hard to see, especially from behind. Teslas are famously equipped only with video cameras for sensors. Other AV makers use LiDAR, which builds a point cloud of surrounding objects by scanning the area using an infrared laser. During a 2018 Tesla earnings call, Tesla’s CEO Elon Musk referenced LiDAR by saying, “In my view, it’s a crutch that will drive companies to a local maximum that they will find very hard to get out of. Perhaps I am wrong, and I will look like a fool. But I am quite certain that I am not.” A year later, he doubled down on that attitude by saying that anyone relying on LiDAR “is doomed.”

Don’t get me wrong, here. Teslas are festooned with video cameras and that’s a very good thing. Seeing its surroundings with the unblinking stare of half a dozen video cameras using a monitoring system that never lets its attention wander should reduce the number of fatal accidents. But visible light has its limitations in the dark, in dust, in rain, in snow, in fog, and in other corner-case situations. 

If you object, saying that LiDAR is too expensive, then consider this: the imaging array on the back of Apple’s iPhone 12 Pro and iPhone 13 Pro smartphones and the Apple iPad Pro have sported LiDAR sensors since 2020. These devices cost a fraction of a product like a Tesla AV. Besides some interesting party tricks, like estimating a person’s height, the LiDAR sensors in Apple products help these devices to take better photos by focusing more accurately in darker and even in well-lit environments.

There’s an interesting bit of additional information about infrared sensing that’s recently come to light (pardon the pun) thanks to the James Webb Space Telescope (JWST). The JWST sees only in the infrared, and its very first pictures reveal many details we’ve not seen from the Hubble Space Telescope’s visible-light images. Infrared light passes through dust and gas better than visible light, so we see more detail in JWST’s images of nebulae. LiDAR generally illuminates scenes with infrared laser light.

Similarly, a properly designed chess-playing robot should always be aware of its opponent’s hand, or anything else within the space that can be occupied by the robotic arm. Just knowing where the chess pieces are on the board is not sufficient in this situation. Perhaps LiDAR is not the solution in this or even many situations, but LiDAR is no longer too expensive. Cost is no longer an excuse.

The chess-playing robot causing havoc in a Moscow chess tournament is an obvious example of a cobot, which Wikipedia defines as “a robot intended for direct human robot interaction within a shared space, or where humans and robots are in close proximity.” By that definition, all AVs including Tesla AVs are cobots as well. In my opinion, any cobot designed without a full awareness of the space it shares with humans, bolstered by more-than-sufficient sensor technology including LiDAR, is an extremely hazardous piece of tech that should be sued out of existence. When humans and cobots try to occupy the same space at the same time, flesh and bone always yield to steel, aluminum, and hard plastic. In a clash between cobot and human, the puny human nearly always loses.

Leave a Reply

featured blogs
Oct 6, 2022
The days of 'throwing it over the wall' are over. Heterogeneous integration is ushering in a new era of silicon chip design with collaboration at its core'”one that lives or dies on seamless interaction between your analog and digital IC and package design teams. Heterogeneo...
Oct 4, 2022
We share 6 key advantages of cloud-based IC hardware design tools, including enhanced scalability, security, and access to AI-enabled EDA tools. The post 6 Reasons to Leverage IC Hardware Development in the Cloud appeared first on From Silicon To Software....
Sep 30, 2022
When I wrote my book 'Bebop to the Boolean Boogie,' it was certainly not my intention to lead 6-year-old boys astray....

featured video

PCIe Gen5 x16 Running on the Achronix VectorPath Accelerator Card

Sponsored by Achronix

In this demo, Achronix engineers show the VectorPath Accelerator Card successfully linking up to a PCIe Gen5 x16 host and write data to and read data from GDDR6 memory. The VectorPath accelerator card featuring the Speedster7t FPGA is one of the first FPGAs that can natively support this interface within its PCIe subsystem. Speedster7t FPGAs offer a revolutionary new architecture that Achronix developed to address the highest performance data acceleration challenges.

Click here for more information about the VectorPath Accelerator Card

featured paper

Algorithm Verification with FPGAs and ASICs

Sponsored by MathWorks

Developing new FPGA and ASIC designs involves implementing new algorithms, which presents challenges for verification for algorithm developers, hardware designers, and verification engineers. This eBook explores different aspects of hardware design verification and how you can use MATLAB and Simulink to reduce development effort and improve the quality of end products.

Click here to read more

featured chalk talk

Prognostic Health Management/Predictive Maintenance Solutions for Machines

Sponsored by Mouser Electronics and Advantech

Prognostic health management, also known as predictive maintenance, is a vital component of our industrial ecosystem. In this episode of Chalk Talk, Amelia Dalton chats with Eric Wang from Advantech about the roles that data acquisition, data processing and artificial intelligence play in prognostic health management, the challenges of building these types of systems, and what kind of predictive maintenance solution would be the best fit for your next design.

More information about Advantech WISE-750 Intelligent Vibration Gateway