Predictive Analytics in Self-Driving Cars

 

Forecasting is the sub-set of predictive analytics. Forecasting has the element of time in the data. Historical time series data is required to build a forecasting model. Combining these two elements, predictive analytics is the basis of self-driving car innovation. The application of prediction to clone human behavior in self-driving cars the common “approach to mimic the actions of human drivers” (Gu, Li, Di, & Shi, 2020, para.1).   

Unlike humans who are relying on previous and current experience, self-driving cars rely on previous and current data to learned from and construct patterns. Gained information are input instructions to safely “drive” cars in the simple and complex car driving environments. Self-driving cars, therefore, are trained using “training data” to predict future events on the road. Self-driving cars use machine learning algorithms that are able to take data from different sources (including reading maps) to predict upcoming road trajectories.

Built using Neural Network algorithm, PredctionNet enables us to gather data. Perception algorithms are based on convolutional neural networks, powered by machine learning, a type of deep learning” (Computer, 2019, p.1). In a recent development for better accuracy of prediction, a newer Neural Network algorithm, ResNet, is developed to increase “the depth of the network by adding residual blocks and reached a stunning depth of 152 layers” (Gu, Li, Di,, & Shi, 2020, para.9).   PredictionNet takes the data from the road and predicts upcoming road interactions. Predicting future events comes with uncertainties. PreductionNet enables to report the uncertainties in statistical terms, which are associated with individual road users.

Blog Post: profdetails.blogspot.com

References

Computer, E. (2019). How to make self-driving cars safer on roads. Express Computer, Retrieved from https://proxy.cecybrary.com/login?url=https://www-proquest-com.proxy.cecybrary.com/trade-journals/how-make-self-driving-cars-safer-on-roads/docview/2199220036/se-2?accountid=144789

Gu, Z., Li, Z., Di, X., & Shi, R. (2020). An LSTM-based autonomous driving model using a waymo open dataset. Applied Sciences, 10(6), 2046. doi:http://dx.doi.org.proxy.cecybrary.com/10.3390/app10062046

Comments

Post a Comment

Popular posts from this blog

Technology and Learning Approches

  The sociotechnical aspects of formal, informal, and non-formal learning approaches in the traditional educational system in Brazil are presented in the research paper. The approaches and meanings of formal/informal/non-formal learning approaches in children’s education require an understanding of the cultural and emotional learnings. Also, these learning approaches require an understanding of other interests and challenges children face in using technology in their learning.   In explaining the research approach, the researchers attempted to address affectiblity, the approaches new technology to be adopted for children to make better use of it and evaluated the integration of new technology in different learning situations, and the implications of the school’s culture on the adoption of new technology in schools. Culture plays an integral role in informal learning (Schürmann & Beausaert, 2016). Although it is widely considered new technology is part of the society, t...
Read more

Serendipity, Error, and Exaptation

                   Serendipity, a happy discovery, is when one can make an unusual connection of ideas to create something new. Serendipity, according to Mirvahedi and Morrish (2017), is “finding and exploring new things is, however, believed to be a result of “accident and sagacity”, rather than accident or luck alone” (para. 5). Serendipity is a micro-learning experience that leads to a groundbreaking discovery (Björneborn, 2017). Steve Jobs, for example, was able to connect the idea between the use of iPhone and the music library. In his serendipity, Steve Jobs created a pathway for innovation. His approach was realized through the approach he was pursuing – not dividing the company’s functions into different areas and all employees to work with each other. His style of leadership led to inventions where he put his engineers have constant contact with each other that lead to serendipitous opportunities, which arise from regular...
Read more