This study highlights the challenge of motion sickness (MS) in autonomous vehicles (AVs), providing a comprehensive review of assessing, predicting, and preventing this issue with a special focus on vehicle dynamics and control-based approaches. Unlike previous studies, this review bridges the gap between MS prediction models and vehicle dynamics-based mitigation strategies by presenting an integrated perspective. Effective mitigation requires accurate and reliable prediction. In this context, motion-based prediction approaches, recognised for their practicality, cost-effectiveness, and promising results, are examined in detail with particular focus on ISO-based methods and sensory conflict theory-based models. The importance of identifying MS triggers and validating these models experimentally is also emphasised, alongside recent trends in customised approaches addressing individual variability in MS susceptibility. The study then investigates mitigation strategies centred on vehicle dynamics and control systems, due to their potential for directly controlling motion triggers, calling for tailored and integrated approaches. Furthermore, the critical role of trajectory planning and tracking algorithms in mitigating MS is reviewed, emphasising their potential through optimal control and the incorporation of MS metrics into cost functions. Additionally, integrating trajectory planning with active chassis systems is identified as a promising direction for reducing MS. The study concludes by underscoring the importance of optimised, personalised, integrated and connected vehicle dynamics and control-based methods to effectively mitigate MS in AVs. Finally, a future horizons approach, supported by a vision roadmap, is introduced as a means to address current challenges, define research directions, and ultimately advance the adoption of AVs with minimum MS.