AWS DeepRacer driving analyzer by Amazon
Recognizing, analysing AWS DeepRacer performance
PROJECT BACKGROUND
To help AWS DeepRacer successfully complete a race, we use the training model powered by Amazon SageMaker. However, the vehicle appears to perform differently when running in reality. Therefore, we are developing a tool that is able to analyze any AWS DeepRacer’s performance through an AI service (by Amazon Rekognition Custom Label).
PROJECT CONCEPT
By analyzing the images or videos, AWS DeepRacer driving analyzer by Amazon Rekognition Custom Label is able to recognize the vehicle’s position on the source should information be available. Useful information such as the speed and the trail can be extracted from the source. Such information is difficult to tell and differentiate by human eyes. In comparison, information generated by AI will be more objective and accurate.