Case Study: Investigating the Performance of Interactive Multiple Motion Model Algorithm for a Crossing Target Scenario
The paper presents a case study on the performance of the interactive multiple motion (IMM) model technique in tracking more than one target, especially when the targets are crossing each other during their motion. We have used a selective approach in choosing multiple motion models, thus providing wider coverage to track straight line as well as manoeuvring targets. Initially, there are two motion models in the system to track each target. The probability of each model being correct is computed through a likelihood function for each. The study presents a simple technique to introduce additional models into the system using deterministic acceleration, which basically defines the dynamics of the system. Therefore, based on this value, more motion models are employed to increase the coverage. Finally, the combined estimate is obtained using posteriori probabilities from different filter models. The case study shows that when targets are well separated from each other and may be manoeuvring, the system adequately tracks them. However, when the targets are close to each other or when they cross each other, the system performance decreases and large tracking error occurs.
KeywordsInteractive Multiple Model (IMM) Target Tracking Simulation Kalman Filter
Unable to display preview. Download preview PDF.
- 1.Bar-Shalom, Y.: Tracking a maneuvering target using input estimation versus the Interacting Multiple Model algorithm. IEEE Transaction Aerospace and Electronic Systems, AES-25, 296–300 (1989)Google Scholar
- 2.Bar-Shalom, Y., Li, X.R.: Estimation and tracking: Principles, Techniques and Software. Artech House (1993)Google Scholar
- 3.Bar-Shalom, Y., Fortmann, T.E.: Tracking and Data Association. Academic Press, Inc., London (1988)Google Scholar
- 4.Blackman, S.S.: Multiple-Target Tracking with Radar Applications. Artech House, Inc. (1986)Google Scholar
- 5.Lin, H.-Y., Atherton, D.P.: Investigation of IMM tracking algorithm for maneuvering target tracking. In: First IEEE Regional Conference on Aerospace Control Systems, pp. 113–117 (1993)Google Scholar
- 6.Hussain, D.M.A.: Motion Model Employment using interacting Motion Model Algorithm. In: Proceedings of International Control Conference (ICC2006), Glasgow, Scotland (2006)Google Scholar
- 11.Hussain, D.M.A.: Tracking multiple objects using modified track-observation assignment weight approach for data association. In: INMIC-2003, International Multi-topic Conference, Islamabad, Pakistan (2003)Google Scholar
- 12.Hussain, D.M.A., Durrant, M., Dionne, J.: Exploiting the Computational Resources of a Programmable DSP Micro-processor (Micro Signal Architecture MSA) in the field of Multiple Target Tracking. In: SHARC International DSP Conference 2001, North Eastern University, Boston, USA (2001)Google Scholar