Many hackers have publicly demonstrated their superb network technology to successfully crack the car system and control the car. But this time, you don’t need any network technology, you only need a projector to fool Tesla - specifically, Tesla’s Autopilot2.5 and Mobileye’s 630 Pro driver assistance system.
A PhD student at the University of the Negev named Ben Nassi discovered a vulnerability and published his research in the paper "ADAS Spectre: Phantom Attacks on Driver Assistance Systems."
In addition, Ben Nassi's team uploaded the test video to Youtube. In the video, they used a projector to project a two-dimensional human image onto the road, and the Model X used for the test did not recognize it as a fake and slowed down.
In another test, the team projected fake lane lines on the ground, and the test vehicle followed the fake lane lines and ignored the real physical lane lines. In addition, the vehicle continued to move along the "lane" at the projected "intersection", resulting in it changing lanes on the actual road.
The research team pointed out that the Model X may have made such mistakes because the virtual lane lines are brighter than the real lane lines.
During the test, Israel's Mobileye 630Pro system also failed to distinguish the virtual image of the projector, causing the test car to not drive according to the actual speed limit of the road.
"Because Tesla and Mobileye are the two top companies in the field of autonomous driving, we will test their products," the research team said.
Tesla is the most outstanding company among many companies developing autonomous driving technology, but autonomous driving has also brought Tesla a lot of lawsuits and troubles. According to the latest news from abroad, according to an investigation, in January 2018, a Tesla Model S crashed into a fire truck parked on the side of the road on a California highway. The fault lies with the improper design of Tesla's autonomous driving system!
The design of Tesla's autopilot system allows drivers to leave the steering wheel, which causes drivers to over-rely on the system. This description of drivers being unable to concentrate is fully responsible for multiple accidents.
Tesla's defense
Tesla has defended the Autopilot system, saying in an email: "Tesla owners have driven billions of miles with Autopilot engaged, and drivers using Autopilot remain safer than those not using assisted driving. Our Autopilot driver monitoring system constantly reminds drivers of their responsibility to stay focused and prohibits using Autopilot if they ignore warnings."
Tesla's car manuals include numerous warnings about the car's automatic following. One section of the manual says it is "not a collision warning or avoidance system." The company also says drivers "should never" rely on Autopilot to slow down.
Several fatal car accidents
Tesla's Autopilot system has been involved in at least three fatal crashes in the U.S., including a fatal crash in Mountain View, California, in March 2018 and another in Delray Beach, Florida, last March. There are also risks from electric vehicles' lithium-ion batteries, which have caught fire after several accidents and are difficult to extinguish.
Traffic safety regulators and the National Transportation Safety Board made safety recommendations in response to the two crashes, while the National Highway Traffic Safety Administration (NHTSA) can order a product recall if it determines a defect poses an unreasonable risk to vehicle safety.
Tesla said in an emailed statement that since the 2018 crash, "we have made updates to the system, including adjusting the time interval between the actual warning and the activation of the warning condition."
It is reported that Tesla is currently developing a more advanced fully automatic driving system, which is planned to be launched next year. During the entire driving process from home to work, the driver does not need to put his hands on the steering wheel, and can even play with his phone or work. However, before that, Autopilot is still just an auxiliary driving system and cannot completely replace humans.
Previous article:Behind the acquisition of Delphi
Next article:Hydrogen fuel cell vehicle technology research and development has made new breakthroughs, and development is gradually entering the "fast lane"
- Popular Resources
- Popular amplifiers
- Detailed explanation of intelligent car body perception system
- How to solve the problem that the servo drive is not enabled
- Why does the servo drive not power on?
- What point should I connect to when the servo is turned on?
- How to turn on the internal enable of Panasonic servo drive?
- What is the rigidity setting of Panasonic servo drive?
- How to change the inertia ratio of Panasonic servo drive
- What is the inertia ratio of the servo motor?
- Is it better for the motor to have a large or small moment of inertia?
Professor at Beihang University, dedicated to promoting microcontrollers and embedded systems for over 20 years.
- LED chemical incompatibility test to see which chemicals LEDs can be used with
- Application of ARM9 hardware coprocessor on WinCE embedded motherboard
- What are the key points for selecting rotor flowmeter?
- LM317 high power charger circuit
- A brief analysis of Embest's application and development of embedded medical devices
- Single-phase RC protection circuit
- stm32 PVD programmable voltage monitor
- Introduction and measurement of edge trigger and level trigger of 51 single chip microcomputer
- Improved design of Linux system software shell protection technology
- What to do if the ABB robot protection device stops
- Wi-Fi 8 specification is on the way: 2.4/5/6GHz triple-band operation
- Wi-Fi 8 specification is on the way: 2.4/5/6GHz triple-band operation
- Vietnam's chip packaging and testing business is growing, and supply-side fragmentation is splitting the market
- Vietnam's chip packaging and testing business is growing, and supply-side fragmentation is splitting the market
- Three steps to govern hybrid multicloud environments
- Three steps to govern hybrid multicloud environments
- Microchip Accelerates Real-Time Edge AI Deployment with NVIDIA Holoscan Platform
- Microchip Accelerates Real-Time Edge AI Deployment with NVIDIA Holoscan Platform
- Melexis launches ultra-low power automotive contactless micro-power switch chip
- Melexis launches ultra-low power automotive contactless micro-power switch chip
- Review summary: National Technology's new M4F core N32G430 is here~
- Can you tell me what the Cf2 capacitor does in the circuit? Thanks!
- What is the purpose of connecting two PMOS in series?
- Top 10 hot technologies in 2019: 5G ranks first, and "brain-computer interaction" achieves breakthrough
- DSP Demo Programming Example Sharing
- Fudan Micro FM33LC046N Evaluation + External Interrupt
- A brief discussion on the technological innovation and market prospects of Wi-Fi 6E
- Why embedded engineers have misunderstandings about 8-bit MCUs
- Disassemble BYD's smart car key to see what solution it uses
- ST official statement: STM8 and STM32 production capacity description