Quantcast
Jump to content
  • Welcome to Auto Parts Forum

    Whether you are a veteran automotive parts guru or just someone looking for some quick auto parts advice, register today and start a new topic in our forum. Registration is free and you can even sign up with social network platforms such as Facebook, Twitter, Google, and LinkedIn. 

     

Researchers at the University of Michigan Develop Neural Networks


Recommended Posts

Chongqing Feilong Jiangli collects the latest information for you, so that you can get the most cutting-edge technical information. Researchers at University of Michigan, by focusing on human gait, body symmetry and foot position, use more precise technology than now to teach autopilot to identify and predict pedestrian actions. Data collected by vehicles through cameras, lidars and GPS allow researchers to capture video clips of human action and reproduce them in three-dimensional computer simulations. On this basis, scientists have created a "biomechanically inspired recurrent neural network" to classify human actions.

According to the researchers, they can use the neural network to predict the movement and future position of one or more pedestrians 50 yards away from the car. If a vehicle wants to have the necessary predictive ability, the neural network needs to study the details of human movements in depth: the rhythm of human gait, the mirror image of symmetrical limbs and the way in which the position of feet affects human stability when walking.

Many machine learning algorithms that upgrade the level of automatic driving to the present level involve two-dimensional images. If a computer is shown millions of pictures with parking signs, it will eventually be able to recognize them in real time. By using a few seconds of video clips, the University of Michigan system can study the first half of the video and make predictions, then confirm the predictions with the second half of the video.

The research results show that this new system improves the ability of the autopilot to predict the most probable situation in the future. Finally, the system can enhance the safety of the self driving vehicle.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

 Share

  • Similar Topics

    • By Erica Zhu Feilong Jiangli
      Chongqing Feilong Jiangli collects the latest information for you, so that you can get the most cutting-edge technical information. A private company focused on developing the first rechargeable lithium-metal batteries in the United States announced that it would cooperate with SKI, Korea's largest energy and chemical company, to develop lithium-metal batteries.
      This cooperation will focus on solid-state lithium anode laminates, which can double the energy density and cycle life of rechargeable batteries. The initial goal was to produce and test prototype batteries to prove whether the increase in volume and weight led to an increase in energy density and cycle life relative to existing lithium-ion batteries.
      SK Innovation plans to complete the research on conductive glass separators by the end of 2021 and apply them to the development of lithium metal batteries. Conductive glass separator is the key technology to promote the stability of lithium metal batteries. The company hopes to successfully develop conductive glass separators to accelerate the commercialization of lithium-metal batteries.
      A battery company based in Berkeley, California, USA, has invented and patented a technology to protect the electrodes of lithium-metal batteries, which is the core technology of sulfur-lithium batteries, lithium-air batteries and water-lithium batteries. Scientists at the battery company have invented a patented process that combines lithium metal with a continuous monolithic glass electrolyte. Battery validation in the laboratory indicates that continuous monolithic glass prevents dendrites and promotes lithium metal recycling.
    • By Erica Zhu Feilong Jiangli
      MINIEYE recently announced that it and Xilinx (Inc) have developed a turnkey sensing system for ADAS. The two companies are working on creating solutions for car companies and first-class suppliers, and running MINIEYE IP on Sales Zynq-7000 and Zynq UltraScale + MPSoC platforms to achieve 1-3 self-driving.
      According to Liu Guoqing, founder and CEO of MINIEYE, the new technology can identify and analyze more than 20 kinds of traffic targets, using the Sellings MPSoC platform. The scheme can realize signal input of cockpit exploration, target geophysical exploration and other sensors. This kind of function is designed to be applicable to complex automatic driving situations. After combining MINIEYE's algorithm with the advanced platform of Sales, the two sides hope that this scheme will help automobile companies and first-class suppliers to reduce R&D costs and improve their R&D efficiency.
      MINIEYE can provide high-precision deep neuron network (DNN), make full use of Sellings vehicle-grade chips, while maintaining low power consumption. MINIEYE has installed the former market products into the major mainstream automobile companies'models.
      This scheme can help users deploy differentiated visual algorithms. The software and hardware of Sellings on-chip system are programmable, and can be equipped with new features and functions for vehicles, aiming at meeting the NACP requirements of different countries.
    • By Erica Zhu Feilong Jiangli
      Farreo has developed an end-to-end learning system for speed control. The system adopts the function of neuron network and long-term memory (LSTM), which is a recurrent neural network (RNN), and can learn long-term dependencies.
      The new system was developed by Wirbel and her colleagues, who also deployed an artificial neural network (ANN), which relies on in-depth learning technology. The network has been provided with a large number of demonstrations of human drivers'vehicle operation, which are recorded by forward-looking cameras and can be closely related to the driver's visual field when driving.
      The neuron network will also be trained to imitate the driver's behavior, especially focusing on reproducing the current speed of the vehicle. For example, when the input image contains a speed-limiting panel of 50 km/h, the neuron network will ensure that the current speed of the vehicle is not higher than 50 km/h.
      Perhaps in the near future, the neuron network system will be deployed in the automatic driving vehicle to improve the efficiency of speed control and achieve more intuitive driving control. Researchers plan to apply this proof-of-concept to more complex driving situations, adding a variety of complex driving manipulations, such as diversion, turning at intersections or vehicle navigation at roundabouts.
    • By Alex
      Open to the public and available for all channels of the aftermarket, Dorman University provides Web-based access to product training related to Dorman’s growing categories and product offering.
       
      Dorman says signing up for its new training is both easy and free. After visiting www.DormanUniversity.com and enrolling in the program, students can begin selecting from a wide variety of available course content including:
       
      · An introduction to Hybrid Drive Batteries
      · Highlights from Dorman’s latest Service Dealer Guide
      · An overview of Dorman’s TPMS program
      · An overview of the diesel parts market
      · Highlights from Dorman’s latest Heavy Duty New Product Guide
      · New training modules added often
       
      Each training module consists of course materials that can be downloaded for easy reference, learning content such as videos or presentations, a brief quiz on the course materials, an opportunity to earn a certificate of completion and a closing review summary.
       
      Enroll today to start participating in Dorman University.
       
      For additional information on Dorman’s latest product introductions, visit www.DormanProducts.com or call 800-523-2492. Also available on the company’s website are links to the Dorman Products signup and links to the company’s Facebook, Twitter and YouTube channels.
       

    • air filtration
    • By APF
      Source: http://www.foxbusiness.com/markets/2017/08/22/lg-to-open-michigan-plant-to-make-electric-car-parts.html
      LG Electronics said Tuesday it will spend $25 million to open a U.S. plant for manufacturing electric vehicle components.
      The 250,000-square-foot building is located in Hazel Park, Michigan, a suburb of Detroit. When it opens next year, the plant will create at least 292 jobs in Hazel Park and an expanded research and development center in nearby Troy.
      The Michigan government is providing a four-year, $2.9 million capital grant for the project.
      “LG’s initiative to develop and produce world-class EV components in the United States represents a key pillar of our strategy to be the best technology partner to U.S. automakers,” said Ken Chang, head of the LG Vehicle Components North American Business Center.
      LG said vehicle components are the company’s fastest-growing business. Auto-related revenue jumped 43% year-over-year to $1.5 billion during the first half of 2017, driven by LG’s supplier agreement with General Motors (GM) for the new Chevrolet Bolt electric vehicle.
×
  • Create New...