Dr Henry Liu, a civil and environmental engineering professor at the University of Michigan, proposed a new testing method for a simulated driving environment, that eliminates bias and improves efficiency.
According to a research paper recently published in Nature Communications by the Center for Connected and Automated Transportation (CCAT), one of the challenges that have hindered the development of autonomous vehicles comes down to a severe inefficiency in the way autonomous vehicle testing and evaluation is performed.
The fizz around autonomous vehicles (AVs) can mostly be attributed to a projected decrease in traffic fatalities and the possibility of providing access to education, healthcare and job opportunities to underserved communities. Unfortunately, the standard test to evaluate the readiness of an AV is still not clear. Currently, state-of-the-art testing combines software simulation, closed-track testing and on-road testing.
The problem with most of the available simulation software and test tracks is that events of interest, including accidents, rarely happen. Thus, systems might require a great amount of money to demonstrate safety performance.
Dr Henry Liu constructed a simulated driving environment using large-scale naturalistic driving data collected by the University of Michigan Transportation Research Institute (UMTRI). This testing method eliminates bias and improves efficiency by training background vehicles to learn when to execute certain adversarial manoeuvres, while simultaneously keeping the environment naturalistic.
In contrast, NADE is allowing for uninterrupted interaction between AVs and multiple background vehicles. For instance, if we want to test our vehicle in an urban environment, this approach would allow the testing AV to drive continuously and experience adversarial scenarios. This environment eliminates the inefficiency of the currently available options by multiple orders of magnitude.
Reuben Sarkar, the president and CEO of the Association for Computing Machinery (ACM) shared that if we drive one mile at ACM using augmented reality simulation with NADE superimposed on our track, it could be equal to thousands of public road miles, leading to dramatically lower overall cost. Sarkar also noted that it is high time to validate AVs in a safer, controlled, repeatable test environment and that ACM sees this as a significant product development resource.