Indoor Tracking Systems (ITS) are becoming increasingly popular due to their affordability and the services they provide, such as indoor navigation and asset tracking. Most of the ITS presented in the literature or deployed in the real world rely on dedicated infrastructures that are costly and difficult to deploy and maintain. An alternative lies in infrastructure-free ITS based on inertial sensors embedded in each mobile device, but such ITS suffer from an accumulation of errors degrading their accuracy. To mitigate this accuracy degradation, an approach consists in fostering collaboration between moving devices. However, a major limitation of this approach lies in the complexity of devising such collaborative inertial-based algorithms. In this paper, we address this limitation by proposing MobiXIM, a framework that unifies the processes of devising, evaluating, and fine-tuning collaborative inertial-based algorithms. In addition, we propose a midpoint algorithm devised using our proposed framework. Through this experiment, we highlight each of the components of MobiXIM that operate separately and communicate with each other, thus allowing researchers to focus solely on developing specific parts of their solutions.