Driving is referred to as a cognitively challenging task that involves key-decision process. However, this decision process can be impeded due to the dynamic changes in the emotions of the driver on the drive, which not only impact their own driving but can also pose a threat to the living elements surrounding the driver which are all a part of a Vehicle Ad-Hoc Network (VANET). Therefore, integration of the affective automotive driver emotion detection devices with the VANETs can improve the dynamic driving experience of all the drivers in the VANETs. This paradigm of driving introduced can help the drivers in the connected networks of the VANETs to not only be vigilant of the physical on-road driving conditions around them, but also helps in the discerning the driving patterns of the co-drivers through communications in the VANETs. Such a system of vigilant driving through emotion detection can result in a cooperative and adaptive driving environment which can benefit the drivers by increasing their safety, reducing the driving-associated risk, and improvising the driver’s driving experience. Thereby enhancing both the driver’s safety and experience on the road. This paper implements the various state-of-art deep learning models in detecting and interpreting human emotions for affective automotive emotion detection systems in VANETs and it has been observed that the SqueezeNet model exhibits the highest accuracy of 99.91%.