Many sports (for example, sailing, multi-racing, various marathons and orienteering) do not allow spectators to constantly watch the athletes. To increase entertainment and control, trackers are used, the data from which is shown to viewers in real time. However, when conducting such online broadcasts, the organizers face the problem of inaccurate display of the position of the athlete. This can make it difficult for viewers to follow the dynamics of the event. The goal of our work is to create a correction system that will correctly display the location of the athlete using a specific model for a particular sport. Our work is devoted to methods of correction for orienteering. To achieve the goal, we implemented and compared several correction methods: naive pulling the athlete to the road, Kalman filter, hybrid options. As a result, one of our hybrid options turned out to be 36% more accurate than the naive one. However, further improvement of this correction algorithm is possible.