An autonomous underwater vehicle (AUV) must use an algorithm to plan its path to distant, mobile offshore objects. Because of the uneven distribution of obstacles in the real world, the efficiency of the algorithm dec...An autonomous underwater vehicle (AUV) must use an algorithm to plan its path to distant, mobile offshore objects. Because of the uneven distribution of obstacles in the real world, the efficiency of the algorithm decreases if the global environment is represented by regular grids with all of them at the highest resolution. The framed quadtree data structure is able to more efficiently represent the environment. When planning the path, the dynamic object is expressed instead as several static objects which are used by the path planner to update the path. By taking account of the characteristics of the framed quadtree, objects can be projected on the frame nodes to increase the precision of the path. Analysis and simulations showed the proposed planner could increase efficiency while improving the ability of the AUV to follow an object.展开更多
We propose an algorithm of embedding ensemble tracking in a stochastic framework to achieve robust tracking performance under partial occlusion,illumination changes,and abrupt motion.It operates on likelihood images g...We propose an algorithm of embedding ensemble tracking in a stochastic framework to achieve robust tracking performance under partial occlusion,illumination changes,and abrupt motion.It operates on likelihood images generated by the ensemble method,and combines mean shift and particle filtering in a principled way,where a better proposal distribution is de-signed by first propagating particles via a motion model,and then running mean shift to move towards their local peaks in the likelihood image.An observation model in the particle filter incorporates global and local information within a region,and an adaptive motion model is adopted to depict the evolution of the object state.The algorithm needs fewer particles to manage the tracking task compared with the general particle filter,and recaptures the object quickly after occlusion occurs.Experiments on two image sequences demonstrate the effectiveness and robustness of the proposed algorithm.展开更多
基金Supported by the National Natural Science Foundation of China under Grant No. 60875071
文摘An autonomous underwater vehicle (AUV) must use an algorithm to plan its path to distant, mobile offshore objects. Because of the uneven distribution of obstacles in the real world, the efficiency of the algorithm decreases if the global environment is represented by regular grids with all of them at the highest resolution. The framed quadtree data structure is able to more efficiently represent the environment. When planning the path, the dynamic object is expressed instead as several static objects which are used by the path planner to update the path. By taking account of the characteristics of the framed quadtree, objects can be projected on the frame nodes to increase the precision of the path. Analysis and simulations showed the proposed planner could increase efficiency while improving the ability of the AUV to follow an object.
基金Project(No.2006AA10Z204)supported by the National High-Tech Research and Development Program(863) of China
文摘We propose an algorithm of embedding ensemble tracking in a stochastic framework to achieve robust tracking performance under partial occlusion,illumination changes,and abrupt motion.It operates on likelihood images generated by the ensemble method,and combines mean shift and particle filtering in a principled way,where a better proposal distribution is de-signed by first propagating particles via a motion model,and then running mean shift to move towards their local peaks in the likelihood image.An observation model in the particle filter incorporates global and local information within a region,and an adaptive motion model is adopted to depict the evolution of the object state.The algorithm needs fewer particles to manage the tracking task compared with the general particle filter,and recaptures the object quickly after occlusion occurs.Experiments on two image sequences demonstrate the effectiveness and robustness of the proposed algorithm.
