Autonomous Vehicle (AV) technology is advancing rapidly, promising a significant shift in road transportation safety and potentially resolving various complex transportation issues. With the increasing deployment of AVs by various companies, questions emerge about how AVs interact with each other and with human drivers, especially when AVs are prevalent on the roads. Ensuring cooperative interaction between AVs and between AVs and human drivers is critical, though there are concerns about possible negative competitive behaviors. This paper presents a multi-stage approach, starting with the development of a single AV and progressing to connected AVs, incorporating sharing and caring V2V communication strategy to enhance mutual coordination. A survey is conducted to validate the driving performance of the AV and will be utilized for a mixed traffic case study, which focuses on how the human drivers will react to the AV driving alongside them on the same road. Results show that using deep reinforcement learning, the AV acquired driving behavior that reached human driving performance. The adoption of sharing and caring based V2V communication within AV networks enhances their driving behavior, aids in more effective action planning, and promotes collaborative behavior amongst the AVs. The survey shows that safety in mixed traffic cannot be guaranteed, as we cannot control human ego-driven actions if they decide to compete with AV.

We present CEMA: Causal Explanations in Multi-Agent systems; a general framework to create causal explanations for an agent's decisions in sequential multi-agent systems. The core of CEMA is a novel causal selection method inspired by how humans select causes for explanations. Unlike prior work that assumes a specific causal structure, CEMA is applicable whenever a probabilistic model for predicting future states of the environment is available. Given such a model, CEMA samples counterfactual worlds that inform us about the salient causes behind the agent's decisions. We evaluate CEMA on the task of motion planning for autonomous driving and test it in diverse simulated scenarios. We show that CEMA correctly and robustly identifies the causes behind decisions, even when a large number of agents is present, and show via a user study that CEMA's explanations have a positive effect on participant's trust in AVs and are rated at least as good as high-quality human explanations elicited from other participants.