Back to Media Center

How To Address Complexities Related To Vehicle Autonomy

How To Address Complexities Related To Vehicle Autonomy

By Sidhart Krishnamurthi, Product Management

According to the SAE Levels of Driving Automation, out of 5 stages of autonomy, we are currently at stage 3 (look at the figure below for reference).

SAE Levels of Driving Automation
SAE Levels of Driving Automation

Whilst we celebrate getting this far, it is important to note that we will remain at this stage until the industry experiences a major technological breakthrough because there is currently no system that has the efficiency to empower a car to drive on its own from point A to point B. Given the fact that legislation lags behind industry progress, a system enables full autonomy is critical — once this is achieved, lawmakers will work on passing relevant legislation, allowing society to realize the benefits of vehicle autonomy.

Delving into why we are stuck at partial autonomy today — the current approach involves repurposing legacy technology such as the GPU for the novel problem of autonomous driving. This is not ideal, as these solutions can only process low-resolution data, leading to an overall lack of accuracy and subsequent lack of safety. As a result, autonomous vehicle (AV) development in the industry is halted at partial autonomy. A purpose-built solution must be developed from the ground up — one that can process high resolution data, paving the way for incredible accuracy under any condition, thus enabling full autonomy.

Recogni’s technology is one such solution. Our product has the ability to process high-frame-rate, high-resolution camera data due to its high computability at a low power consumption. What we are developing will have unmatched object detection accuracy under any circumstance, with capabilities that the human eye cannot achieve. With this, we can enable a car to drive on its own under any condition, with no human intervention. Traditional car makers must realize this, and integrate our product into their self-driving systems to be competitive and profitable in the long run.

In addition to consistent innovation, the industry can address these complexities by providing transparency to the public about where we truly are with regards to autonomy and what it will take to get there. As a pioneer in this space, we at Recogni will espouse this with our podcast and future blog posts that delve into the details of the technology — people are indubitably more reassured when hearing from AI experts themselves! We truly want full confidence in autonomy to prevail.


You May Also Like

3rd Party Media
3rd Party Media
Articles
Converting a network for the fastest inference chip on the market

Recogni Core-Tech Team *Conversion: Making a full-precision network compatible with our chip Introduction Deep Neural Networks (DNNs), in particular convolutional ones, are the go-to approach for visual perception tasks that…

Subscribe to Our Updates

Join our newsletter to stay up to date with Recogni!

Enabling ADAS to Full Autonomy