Giving up autonomous driving is also a kind of reconciliation
1 From glory to bottom, where will autonomous driving go?
The autonomous driving industry has entered a stage of rapid development since 2013. In 2016, the autonomous driving industry ushered in a rapid development During the development stage, investment and financing events by related companies gradually increased, and autonomous driving companies began to bloom everywhere. It reached its peak in 2018, with as many as 472 newly registered companies, 78 comprehensive investment and financing events, and the disclosed investment and financing amount as high as 81.1 billion yuan.
Beginning in 2019, the autonomous driving industry began to enter a stage of steady development. In 2020, due to uncontrollable factors, the pace of development of the autonomous driving industry began to slow down, but the amount of investment and financing in the autonomous driving industry Still as high as 43.63 billion yuan, a year-on-year increase of 136.9%. The autonomous driving track is still hot in 2021. In the first three quarters alone, there have been 69 investment and financing events, with a total investment amount of 62.9 billion yuan. As of October 2022, there are 5,682 autonomous driving companies in my country. In the first half of 2022, there will be 201 new autonomous driving-related companies. More and more players hope to get a share of the autonomous driving industry. Car companies and Internet companies have also come to the same track and started a competitive model. The autonomous driving industry is showing a diversified trend.
Looking back at the self-driving industry before 2022, the development of the self-driving industry seems to have entered a situation where a hundred schools of thought are contending. However, entering 2023, the self-driving industry seems to have been torn into a big hole. The development situation has been broken again and again. At the beginning of 2023, autonomous truck upstart Embark declared bankruptcy. As a unicorn company with a market value of US$5.2 billion, it only took 16 months from its peak to bankruptcy. Waymo, as a recognized benchmark company in the field of autonomous driving, also announced the launch of layoff plans in early 2023. On May 18, TuSimple, a self-driving truck company that is facing a delisting crisis, also announced that its board of directors approved a "further restructuring plan" on May 18, which will eliminate 300 jobs, accounting for 10% of the company's total. 30% of employees; also in May, AIWAYS, a new domestic car-making force, frequently broke news about wage arrears and social security suspensions. Before that, WM Motor had already broken news about work suspensions and store closures. ; Pony.ai also broke the news of laying off 50% of its employees.
Entering the autonomous driving industry in 2023, a series of hot words such as layoffs, bankruptcies, and plummeting market value are constantly lingering. The autonomous driving industry seems to have passed the top of the mountain and is reaching the bottom. The road keeps going crazy, and there still doesn’t seem to be an accurate time or answer for when the bottom can be reached. Whether it is a car company or an Internet company, it seems that there is no persistence and enthusiasm for autonomous driving as in previous years.
2 The road to profitability is bleak, and it may be the original sin to make profits difficult
The "cold wave" in the autonomous driving industry will continue, except that it is all In autonomous driving companies, they are reducing personnel expenses and hoping to survive the cold winter. Many Internet companies that are determined to deploy the autonomous driving industry are also cutting back on autonomous driving investments and preparing to use more resources on other projects. The reason may be that autonomous driving has a long road to profitability, and when the market environment is not ideal, cutting off the tail to survive may be the only option.
The autonomous driving industry is like a bottomless pit, and the funds invested are completely out of proportion to the actual profits that can be obtained. With the development of the autonomous driving industry so far, the actual places where you can experience it may only be in a relatively closed environment such as a few factories and schools, and more functions are still used to undertake tasks such as picking up and delivering express delivery, sightseeing at scenic spots, etc., and these usage scenarios Autonomous driving in today's world can only be regarded as low-speed autonomous driving. There is still a long technical gap between the high-speed autonomous driving scenarios we have been pursuing. Even though low-speed autonomous driving is relatively mature, the market demand and market size are not large, making it unrealistic to achieve large-scale profits.
In fact, car companies and companies involved in the autonomous driving industry have always hoped to apply autonomous driving in high-speed autonomous driving scenarios. This is also the significance of autonomous driving technology research. On June 27, 2020, Didi launched an autonomous driving manned application project in Jiading, Shanghai, hoping to bring the use of high-speed autonomous driving into our lives. However, due to the immaturity of the technology and the high investment costs, Didi’s self-driving taxis have still not achieved large-scale application three years later. The autonomous driving craze has been going on for nearly 10 years, but its technology has not achieved large-scale breakthroughs.
From a technical perspective, due to the shortcomings of autonomous driving perception hardware and deep learning, the development of high-speed autonomous driving is now inseparable from the support of high-precision maps, and because high-precision map detection involves traffic environment data Information, the country has not opened up large areas, which has resulted in autonomous driving based on high-precision maps can only be implemented in some areas. This is like locking autonomous driving in a cage, making it technically impossible to achieve large-scale development. and breakthroughs. Even though many car companies are now pursuing the technical direction of "emphasis on perception, light on maps", and many experts have even proposed the development model of intelligent network connection, it is still in its early stages due to its high technical difficulty and high investment cost.
From a cost perspective, even if high-precision map detection can be implemented on a large scale, high-precision maps need to be updated in a timely manner, which requires a lot of labor costs. This is why at this stage One of the reasons why most car companies are pursuing "emphasis on perception and light on maps". And because the cost of autonomous driving hardware is too high at this stage, if you want consumers to truly enjoy the autonomous driving function, the cost of a single vehicle is beyond imagination. It is not realistic for consumers to buy it on a large scale, which has caused many companies to A lot of investment has been made in autonomous driving, but the profit is very small.
3 Give up or persist, maybe the market will give the answer
Should we continue to insist on autonomous driving? Maybe the market has already given the answer. "Laziness" has always been the driving force for technological progress. When traveling by car, the driver's role will be highly bound. Freeing the driver's hands and allowing self-driving cars to carry passengers and cargo will undoubtedly liberate the necessity of the human role in the travel environment. . However, since the development of autonomous driving, many high-speed autonomous driving experiences have continued to appear. However, people's attitude towards autonomous driving is more of a taste of early adopters. If they want it to truly enter people's lives, many people may still raise questions and concerns.
Currently, autonomous driving is still at the L2 level. In order to highlight the advancement of their technology, many new car manufacturers will describe it as L2 or even L2 in their sales promotions. L3 level autonomous driving undoubtedly provides a wrong signal to the development of autonomous driving. Many autonomous driving accidents are caused by drivers overly trusting the autonomous driving system. Since autonomous driving is a very new thing for consumers at this stage, every accident involving autonomous driving will receive a lot of publicity and popularity, which has also caused everyone's distrust of autonomous driving to continue to expand.
The immaturity of the technology, the low cost, and the distrust of the market have put the development of autonomous driving into an endless loop. This has also led to the lack of large-scale autonomous driving so far. Reasons for application. In order to solve these problems, more time and cost investment are required, which is a challenge for both car companies and Internet companies. The front line for the implementation of autonomous driving is constantly being stretched, making it difficult for many companies to bear the burden and can only give up. Investing in autonomous driving related projects, for them, giving up autonomous driving may also be a kind of reconciliation for themselves. Even so, many people will continue to follow the development path of the autonomous driving industry, but it is still impossible to say when we will see the light of day!
The above is the detailed content of Giving up autonomous driving is also a kind of reconciliation. For more information, please follow other related articles on the PHP Chinese website!

Hot AI Tools

Undresser.AI Undress
AI-powered app for creating realistic nude photos

AI Clothes Remover
Online AI tool for removing clothes from photos.

Undress AI Tool
Undress images for free

Clothoff.io
AI clothes remover

AI Hentai Generator
Generate AI Hentai for free.

Hot Article

Hot Tools

Notepad++7.3.1
Easy-to-use and free code editor

SublimeText3 Chinese version
Chinese version, very easy to use

Zend Studio 13.0.1
Powerful PHP integrated development environment

Dreamweaver CS6
Visual web development tools

SublimeText3 Mac version
God-level code editing software (SublimeText3)

Hot Topics



Written above & the author’s personal understanding Three-dimensional Gaussiansplatting (3DGS) is a transformative technology that has emerged in the fields of explicit radiation fields and computer graphics in recent years. This innovative method is characterized by the use of millions of 3D Gaussians, which is very different from the neural radiation field (NeRF) method, which mainly uses an implicit coordinate-based model to map spatial coordinates to pixel values. With its explicit scene representation and differentiable rendering algorithms, 3DGS not only guarantees real-time rendering capabilities, but also introduces an unprecedented level of control and scene editing. This positions 3DGS as a potential game-changer for next-generation 3D reconstruction and representation. To this end, we provide a systematic overview of the latest developments and concerns in the field of 3DGS for the first time.

Yesterday during the interview, I was asked whether I had done any long-tail related questions, so I thought I would give a brief summary. The long-tail problem of autonomous driving refers to edge cases in autonomous vehicles, that is, possible scenarios with a low probability of occurrence. The perceived long-tail problem is one of the main reasons currently limiting the operational design domain of single-vehicle intelligent autonomous vehicles. The underlying architecture and most technical issues of autonomous driving have been solved, and the remaining 5% of long-tail problems have gradually become the key to restricting the development of autonomous driving. These problems include a variety of fragmented scenarios, extreme situations, and unpredictable human behavior. The "long tail" of edge scenarios in autonomous driving refers to edge cases in autonomous vehicles (AVs). Edge cases are possible scenarios with a low probability of occurrence. these rare events

0.Written in front&& Personal understanding that autonomous driving systems rely on advanced perception, decision-making and control technologies, by using various sensors (such as cameras, lidar, radar, etc.) to perceive the surrounding environment, and using algorithms and models for real-time analysis and decision-making. This enables vehicles to recognize road signs, detect and track other vehicles, predict pedestrian behavior, etc., thereby safely operating and adapting to complex traffic environments. This technology is currently attracting widespread attention and is considered an important development area in the future of transportation. one. But what makes autonomous driving difficult is figuring out how to make the car understand what's going on around it. This requires that the three-dimensional object detection algorithm in the autonomous driving system can accurately perceive and describe objects in the surrounding environment, including their locations,

StableDiffusion3’s paper is finally here! This model was released two weeks ago and uses the same DiT (DiffusionTransformer) architecture as Sora. It caused quite a stir once it was released. Compared with the previous version, the quality of the images generated by StableDiffusion3 has been significantly improved. It now supports multi-theme prompts, and the text writing effect has also been improved, and garbled characters no longer appear. StabilityAI pointed out that StableDiffusion3 is a series of models with parameter sizes ranging from 800M to 8B. This parameter range means that the model can be run directly on many portable devices, significantly reducing the use of AI

Trajectory prediction plays an important role in autonomous driving. Autonomous driving trajectory prediction refers to predicting the future driving trajectory of the vehicle by analyzing various data during the vehicle's driving process. As the core module of autonomous driving, the quality of trajectory prediction is crucial to downstream planning control. The trajectory prediction task has a rich technology stack and requires familiarity with autonomous driving dynamic/static perception, high-precision maps, lane lines, neural network architecture (CNN&GNN&Transformer) skills, etc. It is very difficult to get started! Many fans hope to get started with trajectory prediction as soon as possible and avoid pitfalls. Today I will take stock of some common problems and introductory learning methods for trajectory prediction! Introductory related knowledge 1. Are the preview papers in order? A: Look at the survey first, p

Original title: SIMPL: ASimpleandEfficientMulti-agentMotionPredictionBaselineforAutonomousDriving Paper link: https://arxiv.org/pdf/2402.02519.pdf Code link: https://github.com/HKUST-Aerial-Robotics/SIMPL Author unit: Hong Kong University of Science and Technology DJI Paper idea: This paper proposes a simple and efficient motion prediction baseline (SIMPL) for autonomous vehicles. Compared with traditional agent-cent

Written in front & starting point The end-to-end paradigm uses a unified framework to achieve multi-tasking in autonomous driving systems. Despite the simplicity and clarity of this paradigm, the performance of end-to-end autonomous driving methods on subtasks still lags far behind single-task methods. At the same time, the dense bird's-eye view (BEV) features widely used in previous end-to-end methods make it difficult to scale to more modalities or tasks. A sparse search-centric end-to-end autonomous driving paradigm (SparseAD) is proposed here, in which sparse search fully represents the entire driving scenario, including space, time, and tasks, without any dense BEV representation. Specifically, a unified sparse architecture is designed for task awareness including detection, tracking, and online mapping. In addition, heavy

Target detection is a relatively mature problem in autonomous driving systems, among which pedestrian detection is one of the earliest algorithms to be deployed. Very comprehensive research has been carried out in most papers. However, distance perception using fisheye cameras for surround view is relatively less studied. Due to large radial distortion, standard bounding box representation is difficult to implement in fisheye cameras. To alleviate the above description, we explore extended bounding box, ellipse, and general polygon designs into polar/angular representations and define an instance segmentation mIOU metric to analyze these representations. The proposed model fisheyeDetNet with polygonal shape outperforms other models and simultaneously achieves 49.5% mAP on the Valeo fisheye camera dataset for autonomous driving
