A novel robot navigation algorithm with global path generation capability is presented. Local minimum is a most intractable but is an encountered frequently problem in potential field based robot navigation.Through ap...A novel robot navigation algorithm with global path generation capability is presented. Local minimum is a most intractable but is an encountered frequently problem in potential field based robot navigation.Through appointing appropriately some virtual local targets on the journey, it can be solved effectively. The key concept employed in this algorithm are the rules that govern when and how to appoint these virtual local targets. When the robot finds itself in danger of local minimum, a virtual local target is appointed to replace the global goal temporarily according to the rules. After the virtual target is reached, the robot continues on its journey by heading towards the global goal. The algorithm prevents the robot from running into local minima anymore. Simulation results showed that it is very effective in complex obstacle environments.展开更多
In the process of urbanization, the rapid growth of the number of motor vehicles aggravates the problem of parking difficulties, especially in the parking space, vehicles are not accurate positioning conditions in the...In the process of urbanization, the rapid growth of the number of motor vehicles aggravates the problem of parking difficulties, especially in the parking space, vehicles are not accurate positioning conditions in the parking field of the blindness prominent, the occupancy of special parking space is common, the vehicle in the parking behavior of the "last meter" of low efficiency. In this paper, MongoDB non-relational database and Dijkstra algorithm are used to model the location information to realize the determination of the optimal parking space based on real-time parking space in the field. By calling the Baidu API image recognition interface and the software and hardware interaction of the variable intelligence board, the target vehicle personalized path can be accurately guided. After the vehicle is placed in the vehicle, the identification of the parking behavior and the parking assistance for some vehicles are realized through the geomagnetic induction technology and intelligent video tracking and detection technology. It provides a new way of thinking to solve the end problem of "the last meter" in the difficulty of parking and realize the precise and standardized management of parking Spaces in the parking lot.展开更多
An extended social force model with a dynamic navigation field is proposed to study bidirectional pedestrian movement. The dynamic navigation field is introduced to describe the desired direction of pedestrian motion ...An extended social force model with a dynamic navigation field is proposed to study bidirectional pedestrian movement. The dynamic navigation field is introduced to describe the desired direction of pedestrian motion resulting from the decision-making processes of pedestrians. The macroscopic fundamental diagrams obtained using the extended model are validated against camera-based observations. Numerical results show that this extended model can reproduce collective phenomena in pedestrian traffic, such as dynamic multilane flow and stable separate-lane flow. Pedestrians＇ path choice behavior significantly affects the probability of congestion and the number of self-organized lanes.展开更多
文摘A novel robot navigation algorithm with global path generation capability is presented. Local minimum is a most intractable but is an encountered frequently problem in potential field based robot navigation.Through appointing appropriately some virtual local targets on the journey, it can be solved effectively. The key concept employed in this algorithm are the rules that govern when and how to appoint these virtual local targets. When the robot finds itself in danger of local minimum, a virtual local target is appointed to replace the global goal temporarily according to the rules. After the virtual target is reached, the robot continues on its journey by heading towards the global goal. The algorithm prevents the robot from running into local minima anymore. Simulation results showed that it is very effective in complex obstacle environments.
文摘In the process of urbanization, the rapid growth of the number of motor vehicles aggravates the problem of parking difficulties, especially in the parking space, vehicles are not accurate positioning conditions in the parking field of the blindness prominent, the occupancy of special parking space is common, the vehicle in the parking behavior of the "last meter" of low efficiency. In this paper, MongoDB non-relational database and Dijkstra algorithm are used to model the location information to realize the determination of the optimal parking space based on real-time parking space in the field. By calling the Baidu API image recognition interface and the software and hardware interaction of the variable intelligence board, the target vehicle personalized path can be accurately guided. After the vehicle is placed in the vehicle, the identification of the parking behavior and the parking assistance for some vehicles are realized through the geomagnetic induction technology and intelligent video tracking and detection technology. It provides a new way of thinking to solve the end problem of "the last meter" in the difficulty of parking and realize the precise and standardized management of parking Spaces in the parking lot.
基金Acknowledgements This work was supported by the National Natural Science Foundation of China （Grant Nos. 11202175, 11275186, 91024026, and FOM2014OF001）, the Research Foundation of Southwest University of Science and Technology （No. 10zx7137）, and a Singapore Ministry of Education Research Grant （Grant No. MOE 2013-T2-2-033）.
文摘An extended social force model with a dynamic navigation field is proposed to study bidirectional pedestrian movement. The dynamic navigation field is introduced to describe the desired direction of pedestrian motion resulting from the decision-making processes of pedestrians. The macroscopic fundamental diagrams obtained using the extended model are validated against camera-based observations. Numerical results show that this extended model can reproduce collective phenomena in pedestrian traffic, such as dynamic multilane flow and stable separate-lane flow. Pedestrians＇ path choice behavior significantly affects the probability of congestion and the number of self-organized lanes.
