As part of its 2030 Vision strategy announced in June, Honda aims to have a car with Level 4 self-driving capability on sale by the year 2025. The target date is later than what rivals are promising (typically around 2021) but is probably more realistic, perhaps not because of the ability of the technology but rather an accommodative regulatory environment.

As is becoming the standard practice in the industry, Honda is teaming up with firms specializing on various aspects of self-driving cars. On Thursday, the automaker announced it has signed a 5-year joint research and development agreement with China’s SenseTime, an IT firm specializing in artificial intelligence.

AI is seen as the holy grail for fully self-driving cars that can operate in virtually any set of road conditions a human could. The key is deep learning (also referred to as machine learning), a form of AI where computers “learn” how to perform actions on their own based on given data, i.e. without the need to be explicitly programmed. For self-driving cars, the AI systems are refined by racking up millions of test miles.

According to Honda, SenseTime excels in image recognition technologies, especially recognition of moving objects, powered by deep learning technology. In their new partnership, Honda will join its AI algorithms for environment understanding, risk prediction and action planning (see definitions below) with SenseTime’s moving object recognition technologies.

The goal is to develop a reliable self-driving system that will be able to handle both highways and complex urban environments. However, Honda says the technology can also benefit developments in robotics.

As mentioned, Honda aims to have a Level 4 self-driving car on sale by 2025. This would allow a driver to take their eyes off the road and in some cases even nap behind the wheel. However, in certain situations, such as rough weather or changed road conditions due to an accident or road works, the driver will need to take over after being prompted well in advance. Should the driver fail to take over control, the car will be able to safely stop by the side of a road. The ultimate goal is a Level 5 car where no driver is required.

  • Environment understanding: Estimating the driving environment and the behaviors and intentions of pedestrians and vehicles.
  • Risk prediction: Predicting the future position of pedestrians and vehicles based on the results of estimating the driving environment and the intentions of pedestrians and vehicles.
  • Action planning: Deciding on the actions taken by the vehicle such as stopping, starting and avoiding, and then generating driving trajectory based on the results of risk prediction.