Nerd For Tech
Published in

Nerd For Tech

Analysis of the Four Key Technologies of Autonomous Driving — Part 2

Route Plan

Once the intelligent vehicle has a driving task, the path planning of the intelligent vehicle is to find a passable path according to a certain search algorithm, based on perceiving environmental information and determining the vehicle’s position in the environment, thereby realizing the autonomous navigation of the intelligent vehicle. The path planning method can be divided into two categories according to the completeness of the working environment information of the intelligent vehicle:

A Global Path Planning Method Based on Complete Environmental Information

For example, there are many roads from Shanghai to Beijing, and the overall plan is to map out a driving route. Such as the grid method, visual graph method, topology method, free space method, neural network method, and other static path planning algorithms.

A Local Path Planning Method Based on Real-time Sensor Acquisition of Environmental Information

For example, vehicles or obstacles will be on the globally planned route from Shanghai to Beijing. If you want to avoid these obstacles or vehicles, you need to turn and adjust the lane. This is local path planning. Local path planning methods include artificial potential field method, vector field histogram method, virtual force field method, genetic algorithm, and other dynamic path planning algorithms.

Sport Control

After planning the driving path, the next step is to control the vehicle to travel along the desired trajectory, which the motion control part needs to accomplish.

Motion control includes lateral control and longitudinal control. In short, lateral control is steering control, and longitudinal control is speed control. Now more research is on lateral control, and the methods used mainly include synovial control, fuzzy control, neural network control, optimal control, adaptive control, and pure tracking control, etc.

Generally speaking, the lateral control sets a speed, and the vehicle is driving along a predetermined trajectory by its control of the steering. The purpose of longitudinal control is to meet the speed requirements during the driving process, and sometimes it is necessary to cooperate with the lateral control to ensure the safety, stability, and comfort of the vehicle in the course of tracking. Because the vehicle is a particularly complex system, and coupling relationship exists in transverse, so longitudinal, and vertical directions, intelligent vehicles should be controlled transversely, longitudinally, and vertically or even in a coordinated fashion. Because of the complexity of its coupling relationship, the cooperative control technology of intelligent vehicle motion control is also the technical difficulty of this part.

If car companies and tech companies can make these four technologies work together perfectly, then cars can drive themselves, and transportation will usher in the era of autonomous driving.

The Demand for Data Labeling Continues to Increase

The mainstream algorithm model of autonomous driving is mainly based on supervised deep learning. It is an algorithm model that derives the functional relationship between known variables and dependent variables. A large amount of structured labeled data is required to train and tune the model.

On this basis, if you want to make self-driving cars more “intelligent”, and form a closed loop of the business model for self-driving applications that can be replicated in different vertical landing scenarios, the model needs to be supported by massive and high-quality real road data.

In the field of autonomous driving, data annotation scenes usually include changing lanes and overtaking, passing intersections, unprotected left and right turn without traffic light control, and some complex long-tail scenes such as vehicles running red lights, pedestrians crossing the road, and roadsides as well as illegally parked vehicles, etc.

The current artificial intelligence is also called data intelligence. At this stage of development, the more layers of the neural network, the larger amount of labeled data is needed.

For deep learning, data is meaningful only if it is well labeled.


Outsource your data labeling tasks to ByteBridge, you can get the high-quality ML training datasets cheaper and faster!

  • Free Trial Without Credit Card: you can get your sample result in a fast turnaround, check the output, and give feedback directly to our project manager.
  • 100% Human Validated
  • Transparent & Standard Pricing: clear pricing is available(labor cost included)

Why not have a try?




Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store



Data labeling outsourced service: get your ML training datasets cheaper and faster!—