Target Location and Identity Estimation and Fusion using Disparate Sensor Data
Parthsarathy Naidu, Mohammed Mudassar, G Girija, Jitendra R. Raol · AIAA Guidance, Navigation, and Control Conference and Exhibit · 2005
A scheme for estimation of target identity and location using synthesized data of disparate sensors is presented in this paper. Target scenario generation is carried out using marked spatial point process. Infrared sensor data and acoustic sensor data for the scenario are simulated using sensor mathematical models. The sensor data are fused using Bayesian fusion wherein the likelihood functions of the two sensors are used to obtain the posterior probabilities. From the posterior probability distribution, the target identity and location are established using search algorithms. A comparison of four search algorithms, Metropolis Hastings, Simulated annealing, Gradual greedy and Ant colony optimization is made for a two dimensional scenario with two targets. Comparative performance is studied based on several criteria. It was seen that gradual greedy algorithm has a better performance compared to the Metropolis Hastings and Simulated annealing algorithms. The ant colony optimization algorithm has performance which compares with the gradual greedy algorithm and may be preferred for real time applications since it is amenable for parallel implementation.