The computing device 202 may be used to operate an adaptive rule-based system 312, such as an adaptive learning system described below with respect to FIG. 4. The adaptive rule-based system 312 may use a continuous data stream or, in some examples, the adaptive rule-based system 312 may be triggered based on changes in the inputs or conditions provided as inputs to the adaptive rule-based system 312. The adaptive-rule based system 312 may rely on statistical learning methods to trigger application of the adaptive rule-based system 312 to process only the most current data gathered. In some examples, a change in trip status or a change in traffic conditions may prompt an action by the adaptive rule-based system 312. In some examples, the adaptive rule-based system may operate continually or at set intervals, such as every few second, every few minutes, or any other suitable period of time. The adaptive rule-based system 312 may be an adaptive learning system. An adaptive learning system is type of machine learning model intended to simulate the human brain and nervous system. Generally, an adaptive learning system represents a network of interconnected nodes, similar to a biological adaptive learning system, where knowledge about the nodes is shared across output nodes and knowledge specific to each output node is retained. Each node represents a piece of information. Knowledge can be exchanged through node-to-node interconnections and node-to-task connections. Input to the adaptive learning system activates a set of nodes. In turn, this set of node activates other nodes, thereby propagating knowledge about the input. At each set of nodes, transformative functions may be applied to the data. This activation process is repeated across other nodes until an output node is selected and activated. A convolution adaptive learning system (CNN) is a type of adaptive learning system that can exploit the spatial structure of data (e.g. audio files) to classify the data. To do this, a CNN may include one or more of a convolutional layer, a fully connected layer, and a pooling layer.