Most devices on an IoT network have severe limitations on computing resources in terms of memory, processing, and power, which makes these networks generally known as Low Power and Lossy Networks (LLNs). Device mobility in these scenarios has several negative impacts on the network, particularly in terms of connectivity, as node movement affects the routes established by the routing protocol for message routing. The performance of a network is highly related to the adjacent routing protocol. The protocol considered as standard for IoT networks was designed for static networks, and it has a low reactivity in mobility scenarios, besides other limitations. Lightweight On-demand Ad hoc Distance-vector Routing - Next Generation (LOADng) is currently the primary reactive protocol for LLNs. Considering that mobility is a crucial feature for the future of the Internet of Things, this paper proposes EKF-LOADng, a solution built from LOADng for LLNs in the mobility context. The solution aims to make the mobile nodes aware of its location, as well as to predict the nonlinear trajectory of these nodes through the Extended Kalman Filter. Early simulations show that features provided by EKF-LOADng can optimize network performance in terms of flow, connectivity, and latency.