Ghost Locomotion Emerges from Stealth Mode, Plans to Retrofit Cars for Self-driving
【Summary】A new Silicon Valley startup named Ghost Locomotion has emerged from stealth mode. Ghost’s autonomous driving system is designed to retrofit a production vehicle for highway driving, a much less complicated problem for software engineers to solve.
A new Silicon Valley startup named Ghost Locomotion has emerged from stealth mode. The Mountain View Calif-based company joins dozens of other automaker and startups working on self-driving cars.
However, Ghost, which was founded in 2017, is not building self-driving cars, instead its is taking production vehicles and retrofitting them become autonomous capable.
The company mission is to offer self-driving technology to the millions of regular vehicles that are already on the road. Unlike Waymo and Uber, which are designing systems that take people from door-to-door, Ghost's autonomous driving system is designed for highway driving, a much less complicated problem for software engineers to solve.
Ghost raised $63.7 million in capital to date from Founders Fund's Keith Rabois, Khosla Ventures' Vinod Khosla, and Sutter Hill Ventures' Mike Speiser, according to Venturebeat. All three investors have a seat on the company's board of directors.
Ghost bills itself as an AI company that drives your vehicle for you on the highway. The company takes a standard car and adds a hardware kit that includes a suite of cameras to turn it into a self-driving vehicle. Once installed on the vehicle, the tiny cameras are neatly concealed. Ghost says the system works with 2012 model year and newer vehicles, as long as the vehicle is equipped with electric power steering.
Ghost bills itself as an AI company that drives your vehicle for you the highway. The company takes a standard car and adds a hardware kit that includes radar, ultrasonic sensors, as well as a suite of cameras to turn it into a self-driving vehicle. Once installed on the vehicle, the tiny cameras are neatly concealed. Ghost says the system works with 2012 model year and newer vehicles. The vehicles must be equipped with a steer-by-wire system.
By focusing on solving highway driving Ghost is simplifying the enormous technical challenges of self-driving vehicles. The company says it can deliver real self-driving, exit-to-exit on the highway, and solve a problem for everyone driving on today's roads.
Tesla's Autopilot and General Motors' Super Cruise automated driving systems were also developed initially for highway driving.
John Hayes, founder and CEO of Ghost, wrote in a blog post that highway driving is the simplest and most common form of driving today. He wrote that two-thirds of all miles driven in the U.S. are on highways, describing it as being "designed for high speeds with minimal complexity." Whereas self-driving vehicles in urban environments face a much more difficult challenge, having to contend with unpredictable pedestrians, bicyclists and other motorists.
Ghost is designing its software to drive better than the best human driver. Ghost says its technology allows a driver to relax on the highway and fully disengage for their entire commute.
"Most people agree that if a self-driving car performs better than a human driver, it is fit to drive on today's roads. That's why at Ghost, humans are the standard against which we measure ourselves." wrote Hayes in a blog post.
Ghost starts by collecting camera data crowdsourced from drivers. The data give Ghost's engineering team insights of how humans drive in the real world. Ghost uses the data, along with other important mapping data, to train machine learning models to drive more like a human.
To gather as much training data as possible, Ghost recruited Uber and Lyft drivers to help them collect it. Ghost plans to add hundreds of vehicles to its data collecting fleet in the upcoming months.
Ghost is working to ensure that its system operates with the highest level of safety, beginning with software execution, the company says. Ghost claims it's the first self-driving company to formally verify its real-time driving system, a process the company says is usually reserved for the aerospace and defense industries, to ensure perfect software execution. Through formal verification, Ghost says its autonomous driving system is void of any software bugs or glitches.
With tech companies and automakers are busy planning the launch of commercial autonomous robotaxi services and self-driving cars that can navigate in urban environments, automated highway driving just might an easier problem to solve in the interim. Ghost is one of the few companies specifically focused on achieving that goal.
"We believe the highway is the fastest way to get there." Hayes wrote.
Lidar Startup Ouster Introduces a More Affordable 32-Channel Lidar Sensor for Self-driving Cars
China’s BYD Receives Order for 130 Electric Buses, its Biggest Order to Date in the U.S.
BMW Announces its Next-Gen Digital Key Technology
Tesla Unveils the First Model 3 Sedans Made in China
Volvo Cars to Use Blockchain to Track the Source of Cobalt for EV Batteries
Tesla Will Unveil its new Electric ‘Cybertruck’ on Nov 21
Uber Reports $1.2 Billion Loss in Q3, CEO Says the Company Will Be Profitable by 2021
GM Recalls 680,000 SUVs & Pickups Due to Faulty Brake Sensor
- Porsche & Boeing to Develop VTOL Aircraft for Premium Urban Air Travel
- Trump Criticizes Automakers for Not Backing Fuel Economy Rollback
- General Motors CEO Sets Sights on Selling 1 Million EVs Annually
- Volkswagen Says its Not Interested in Buying Tesla Stake
- Audi’s AI:Trail Quattro Concept Is the Autonomous, EV That Looks out of This World
- Continental AG Powertrain Arm Rebrands as Vitesco Technologies, Supplying Electric Drive Units to Hyundai and Groupe PSA
- Electric Car Prices Finally Beginning to Trickle Down
- Elon Musk Confirms Higher Capacity Battery Packs for the Tesla Model S/X Plaid Variants
- Toyota Confirms a Compact Mass-Market EV, Here's What We Know So Far
- Huawei to Develop Radar for Self-Driving Cars in Push into Auto Industry, Execs Say