- 41
- 229 573
DeepRoute
United States
Приєднався 28 гру 2019
Founded in February 2019, DeepRoute.ai is a leading artificial intelligence company dedicated to the research, development and application of autonomous driving technology. As data accumulation and technology iterations continue, DeepRoute.ai aims to create general artificial intelligence in robots, bringing new productivity to humanity.
Smart driving in Chongqing, a 8-dimension city
Smart driving in Chongqing, a 8-dimension city
Переглядів: 191
Відео
The Singularity Moment for Cars
Переглядів 4521 день тому
Maxwell Zhou gave a speech at Global Intelligent Vehicle Summit in Hefei. Here's the highlight video. - We’re developing VLA model based on Vision Transformer with NVIDIA DRIVE Thor. - I think an AI car means it’s human-like and thinks like a human, and it is what consumer are looking for. - Our European operations center will be established to help automakers with overseas deployment. - We hop...
Smart Driving Odyssey - Chengdu, city of pandas
Переглядів 169Місяць тому
The first mass produced vehicle integrated with DeepRoute.ai smart driving solution is officially released! Check out its performance among complex and narrow lanes.
AI, reshaping the automotive industry
Переглядів 954 місяці тому
On June 1, 2024, CEO Maxwell Zhou attended #ManMachineMobility conference and shared his thoughts on AI 2.0 moving forward to AI 3.0, #embodiedAI, and our end-to-end progress. Here's the highlight of the speech.
Key elements and limitations of AGI in robots - GTC speech pt.3
Переглядів 936 місяців тому
Key elements and limitations of AGI in robots - GTC speech pt.3
Explore Scaling Law in Robots - GTC speech pt.2
Переглядів 806 місяців тому
DeepRoute showcases its end-to-end model road test results, and is devoted to find the Scaling Law in robots.
Evolution of DeepRoute.ai Robotics System - GTC speech pt.1
Переглядів 1897 місяців тому
As one of the few AI companies been selected to speak at NVIDIA GTC 2024, DeepRoute.ai shared its robotics system evolution over the past few years, and strategic thinking behind.
NVIDIA GTC 2024 - road test in downtown Shenzhen
Переглядів 1,2 тис.7 місяців тому
DeepRoute.ai CEO Maxwell was speaking at NVIDIA GTC 2024 on end-to-end model, commercialization and future plans with DRIVE Thor. The latest road test in downtown Shenzehn was released during the session. Full details about Maxwell Zhou's GTC session can be accessed here and will be live-streamed for virtual attendees starting from March 23 PST onward. www.nvidia.com/gtc/session-catalog/?tab.al...
DeepRoute.ai Partners with Tencent Map; Unveils Collaboration for 2024 Smart Driving Cars Release
Переглядів 2519 місяців тому
DeepRoute.ai has recently announced collaboration with Tencent in the field of map. The joint solutions will be integrated in consumer vehicles set to debut in 2024, encompassing a range of car models such as sedan, SUV and MPV. This marks the first commercial implementation of both the Driver 3.0 Mapfree solution and Tencent Standard Definition map (SD map).
Driver 3.0 Mapfree Review
Переглядів 19310 місяців тому
During the Guangzhou Auto Show 2023, we invited a few media for a test ride in the car integrated with D-PRO. The trip is around 7km in the city centre.
Driver 3.0 Mapfree Road Test During Guangzhou Auto Show
Переглядів 22711 місяців тому
Driver 3.0 Mapfree Road Test During Guangzhou Auto Show
IAA Mobility Premiere - The latest Driver 3.0 HD Map-Free Road Test in CBD Shenzhen
Переглядів 1,1 тис.Рік тому
Premiere at IAA Mobility show - The car integrated with Driver 3.0 HD Map-Free smart driving solution can handle all kinds of complex scenarios in the urban, highways, tunnel etc.
Urban and Highway Navigation in Shanghai with Driver 3.0 HD Map-Free Solution
Переглядів 858Рік тому
“Embracing the New Era of Automobile Industry” During the Auto Shanghai 2023, our Robotaxi integrated with the latest mapfree solution drove over 20 kilometers along urban roads, expressways and highway, navigating smoothly, safely and efficiently in different scenarios. The latest Driver 3.0 solution will offer brand new riding experience that is more intelligent, more relaxed and more cost-ef...
Car integrated with Driver 3.0 HD Map-Free solution driving in Shanghai during rush hour
Переглядів 3,8 тис.Рік тому
DeepRoute.ai successfully completed HD map-free self-driving public road tests, breaking limitations created by geo-fencing. #fsd
Driverless autonomous vehicle is now official in Shenzhen
Переглядів 191 тис.2 роки тому
www.deeproute.ai/en/news/316.html The Intelligent Connected Vehicles Regulation of Shenzhen's Special Economic Zone comes into force on August 1, 2022. This is the first of its kind regulation in China. Following this enforcement, driverless vehicle is now officially legalized on the public road.
DeepRoute.ai L4 Autonomous Driving Performance in Rainstorm and Snow
Переглядів 2 тис.2 роки тому
DeepRoute.ai L4 Autonomous Driving Performance in Rainstorm and Snow
The Piano Tuner Hails Robotaxi with new accessibility features
Переглядів 5132 роки тому
The Piano Tuner Hails Robotaxi with new accessibility features
The Most Advanced Robotaxis - A Preview of Future Mobility
Переглядів 1,1 тис.2 роки тому
The Most Advanced Robotaxis - A Preview of Future Mobility
Autonomous Trucks to Combat Omicron Variant in Shenzhen
Переглядів 2102 роки тому
Autonomous Trucks to Combat Omicron Variant in Shenzhen
Infrastructure for Autonomous Driving Vehicles
Переглядів 6432 роки тому
Infrastructure for Autonomous Driving Vehicles
Third Year Anniversary Special. The Journal of a Robotaxi.
Переглядів 1162 роки тому
Third Year Anniversary Special. The Journal of a Robotaxi.
German expat family discoverying Robotaxi in Shenzhen
Переглядів 6792 роки тому
German expat family discoverying Robotaxi in Shenzhen
How Perception Works for Autonomous Driving Vehicles
Переглядів 1,5 тис.2 роки тому
How Perception Works for Autonomous Driving Vehicles
Series-Production-Ready L4 Self-Driving Solution at USD 10,000
Переглядів 3,6 тис.2 роки тому
Series-Production-Ready L4 Self-Driving Solution at USD 10,000
The First Asian Female of Hugo Award Winner in Discussion with our CEO
Переглядів 1502 роки тому
The First Asian Female of Hugo Award Winner in Discussion with our CEO
The first self-driving medium duty truck in China
Переглядів 6282 роки тому
The first self-driving medium duty truck in China
First ride with DeepRoute.ai Robotaxi
Переглядів 1,5 тис.3 роки тому
First ride with DeepRoute.ai Robotaxi
Living and Dreaming. I Carry You Forward.
Переглядів 2063 роки тому
Living and Dreaming. I Carry You Forward.
thats not yielding pedestrian
Do you plan on announcing publicly that you stole tesla fsd stack a few years back? If not, how do you expect anyone to buy this? It clearly infringing on tesla intelectual property... When you suddenly multiple companies in china had self driving software that looked exactly like fsd, from every single UI design decision to the behaviour, the whole world should have called you out. You just merged it with your lidar obstacle stack and hd map localisation to make it more robust, which is cool, but its so obvious that the other layers like the lane routing, cam to bev, etc are stolen directly. there is no way you managed to reproduce the visual gllitches and behaviours of the visuals so closly, witout copying some of the achitecture directly. I hope you responde to this, as im a huge fan of what you, Tesla, Wayve, google, comma and so on do, but stealing the old softare from a competitor to use as teh base for your stuff and than advertise it and sell it as "your" product, without even mentioning that you would not exist anymore witout tesla is just wrong.
Do you know about the Apollo open-source project of Chinese company Baidu. Most Chinese autonomous driving companies are built on the framework of Baidu Apollo and have implemented it through engineering. As an autonomous driving algorithm engineer in China, I can responsibly tell you that in the companies I have worked for, our code is based on open source projects and publicly available papers, and we have engineered and implemented them. In the era of fsd v11, Tesla used the bev+transformer route, but it did not have open source code. Shanghai AI lab published a paper called bevformer based on the details of Tesla AI Day and opened source the code, which allowed many Chinese algorithm companies to carry out engineering implementation based on it. At the same time, in addition to Tesla's perception route, many Chinese companies only used the bevpeth technology route and sparse4D technology route, which are different from Tesla. In the era of FSD V12, Tesla used end-to-end technology, but the UniAD paper(2023 cvpr best paper) published by the Chinese company (posted on arxiv in November 2022) was eight months earlier than Musk's first proposal of end-to-end Twitter (in August 2023). At least in my opinion, there are many excellent autonomous driving companies in China with technology routes and details that are different from Tesla, and even have many innovative points, but they are slower than Tesla in terms of engineering implementation.
Nice! yoú even copy pasted the parts of fsd that highlight the area the car wants to go into on a lange change. It looks the exact same just like the rest of the UI and the blue highliting of the area that is target on a lane change even glitches out in exactly the same way as it does on fsd. When you steal their whole software stack, why do you do it so blatantly? This is really sad. Make your own shit guys!
Harmonyspace cockpit
Which car is it?
Please show unedited videos, raw material , or go live !, that would be interesting to watch !. Instead of just 2 minutes... The difference between ... "I want to believe" vs "here let me introduce you the real aliens we've all be waiting for" xD
Propaganda as the usual of the communist
Average vietnamese
Chinese Government and Chinese scientist are great and people are very hard working God Jesus Christ blessed thanks
In China, cars have the right-of-way over pedestrians even if the light is green for both. This car yields to pedestrians and bad drivers.
Does the car really speak ?
It’s amazing that the driverless car is much politely and safely.
in this video you can see how much idiots have a drivers licence ..sensors on this vehicle are wow I have to say.
Praticed million of times with those idot
Very impressive!
This will be a power use behemoth - from the local edge devices to the toroidal mesh of "comms"
Some crazy chinese drivers and cyclists there.
This is fake
Looks great with one car, bc is driving safely and others also react to it when it makes some mistake. How will one AI car react to another AI car? how will AI safedriving/hesitation react to other AI hesitation? How will micro decisions of human vs human compare to AI car to AI car? Looking great nontheless
no such thing as easy or not or give or let or etc, doesn't matter ceptuxuax, do, be, out, can do, be, out etc any nmw and any s perfect
Cool
Awesome
Too bad it still depends on lidar and other expensive sensors. Vision is the solution.
Wrong Tesla made the wrong call 😂
@@alhkcblack9617 Lol funny. My Tesla can already drive itself anywhere.
@@neodark414 it'll drive better with lidar + camera fusion. Lidar will get smaller and smaller, cheaper and cheaper and Tesla will adopt it again later i think.
@@alhkcblack9617 There is still no point. You can get much better resolution with a camera than even the best lidar. Cameras are better, cheaper and more reliable. The only thing stopping everyone else from doing the same thing is that they are years behind on the AI needed to pull it off with cameras.
@@neodark414 you can say it but i won't believe it. I'd rather ride a waymo over a Tesla fsd.
Bullshit #1 anyways. I ask who will go to jail once it kills people
I guess that would explain why waymo and other companies promptly took their robotaxis off the streets. When you start to put all your trust in AI, you become completely dependent on it and when something goes wrong, and it will, no one will know what to do
Badly edited. Fake windscreen.
Good job 1:06
simply stunning and outstanding! i m speechless. no words. 😮 😮
Wow what the hell, is this really level 5 (that is without even safety officer and under any environment)?
the hardest part of algorithm was faking the windscreen! 😂🤣
The reason people want robotaxis is so they DON"T have to hear someone else's conversation! I hope the car doesn't really talk during the ride.
If it's driving is human-like, what's wrong with human driving anyway... Don't expect computers to do things without mistakes and improve every aspect of our lives. Don't put your life in the hands of greedy tech companies. No one's gonna be by your side after a life-changing accident. Think!
better than tesla
但是感觉bev并不准,经常发现车道线横向飘逸,指近处
At two points in the video the behind-view and the in-car view are not the same. Seems like an edit-error.
oh its just the fov
Would like to see more left turns.
My man, you drove great!
NEGARA CHINA MEMANG RAJANYA TEKHNOLOGI-TEKHNOLOGI AI CANGGIH-CANGGIH BANGET DIDUNIA 🎉
It seems like the video has been authored at the border between traffic and front glass. Looks like fake to me
too fast, like terms and conditions in contracts read on a radio ad ... a more raw version might come across as more trustworthy ... and believable ..., no test drivers showing this amazing new technology .. Seems like edited , not live.
Video great. Voice annoying
Is this going to be sold in the United States if so do you need testers?
future is here.
impressive bike collision at the 3-minute-mark ua-cam.com/video/6v036bBD31o/v-deo.html
What irritates me and makes me doubt the video is the fact that this are two different videos fused together. The road view is from a different camera than what we see of the inside of the car and therefor it's tough to judge if all is legitemate here.
im amazed by this tech so many assholes on the road, its so annoying watching them where is police?
It passed a red traffic light at 2:08 ....
highly suspect considering it looks better than Tesla's current state and they won't have fraction of the expertise or training data
The weird synthesized voice and passive aggression in the script are even better than the drive itself.
I hope it keeps learning because the first thing a human is going to do seeing a driverless car is ignore it and walk in front of it or drive in front of it.
Amazing.
This car is causing multiple hazards on it's trip because it's trying to be safe above all else when literally no one near it is driving that way. It's like driving 60 mph on a road where everyone else is going 80. Also, the pedestrians / cyclists in asian cultures have developed habits of weaving through oncoming traffic based on what they think the driver is probably going to do. This car changes all of that and you could see it in the pedestrians hesitations.