# AUTO-DISCERN: Autonomous Driving Using Common Sense Reasoning

**Authors:** Suraj Kothawade, Vinaya Khandelwal, Kinjal Basu, Huaduo Wang, Gopal Gupta  
**arXiv:** [2110.13606](https://arxiv.org/abs/2110.13606)  
**Date:** 2021-10-17  
**Summary page:** https://www.summarizepaper.com/en/paper/2110.13606v1/

## Key points

- The paper explores the use of common sense reasoning in automating driving decisions for autonomous vehicles.
- Machine learning techniques have been widely used for observing and understanding surroundings, but driving decisions are better automated through commonsense reasoning.
- Answer set programming (ASP) and s(CASP) ASP system are introduced as tools to automate commonsense reasoning.
- These technologies can be applied to automate the decision-making process in driving.
- The AUTO-DISCERN system utilizes ASP and s(CASP) to simulate human-style reasoning and make explainable and ethically sound driving decisions.
- The goal is to create an autonomous driving system that mimics the cognitive processes of a human driver.
- Commonsense reasoning enhances safety, reliability, transparency, and accountability in the decision-making process of autonomous vehicles.
- Using appropriate technology for specific tasks is important in autonomous driving.
- Machine learning is effective for perceiving surroundings, while commonsense reasoning is more suitable for automating driving decisions.

## Summary

The paper titled "AUTO-DISCERN: Autonomous Driving Using Common Sense Reasoning" explores the use of common sense reasoning in automating driving decisions for autonomous vehicles. The authors argue that while machine learning techniques have been widely used in autonomous driving technology for observing and understanding the surroundings, driving decisions are better automated through commonsense reasoning. The paper introduces answer set programming (ASP) and the goal-directed s(CASP) ASP system as tools to automate commonsense reasoning. The authors discuss how these technologies can be applied to automate the decision-making process in driving. They present the development of the AUTO-DISCERN system, which utilizes ASP and s(CASP) to simulate human-style reasoning and make explainable and ethically sound driving decisions. The goal of this research is to create an autonomous driving system that mimics the cognitive processes of a human driver. By leveraging commonsense reasoning, the system aims to ensure correct decision-making as long as the modeling and inputs are accurate. This approach not only enhances the safety and reliability of autonomous vehicles but also provides transparency and accountability in their decision-making process. Overall, this paper highlights the importance of using appropriate technology for specific tasks in autonomous driving. While machine learning is effective for perceiving surroundings, commonsense reasoning is deemed more suitable for automating driving decisions. The proposed AUTO-DISCERN system demonstrates how ASP and s(CASP) can be leveraged to achieve this goal, paving the way for more advanced and reliable autonomous vehicles in the future.