TY - GEN
T1 - Study on UGV path selection method based on GIS database
AU - Wang, Meiling
AU - Wang, Xinping
N1 - Publisher Copyright:
© 2014 IEEE.
PY - 2014
Y1 - 2014
N2 - As an important research field of ITS, UGV will improve the efficiency of the urban traffic system by its high intelligence. GIS, which has strong capabilities of storage and management for geographic information, will play a key role in path planning for UGV. Firstly, construct a GIS database aiming at the needs for perceiving the urban traffic environment and storing traffic information related to the perception and decision-making of UGV; secondly, establish a prediction model of travel time based on RBF neural network to predict the travel time of each road section, then use the actual travel time achieved by real car experiments to constantly update the predictive travel time; thirdly, a path selection method based on probability is proposed according to people's demands of daily trips; lastly, the GIS database in the simulation will provide several alternative paths, with which the probabilities of arriving on time and arriving at the optimal time can be calculated for each path, helping UGV select path in accordance with the actual driving tasks. Simulations shown that the proposed path selection method based on GIS database can provide more efficient paths for UGV driving, guiding UGV to complete the driving tasks successfully.
AB - As an important research field of ITS, UGV will improve the efficiency of the urban traffic system by its high intelligence. GIS, which has strong capabilities of storage and management for geographic information, will play a key role in path planning for UGV. Firstly, construct a GIS database aiming at the needs for perceiving the urban traffic environment and storing traffic information related to the perception and decision-making of UGV; secondly, establish a prediction model of travel time based on RBF neural network to predict the travel time of each road section, then use the actual travel time achieved by real car experiments to constantly update the predictive travel time; thirdly, a path selection method based on probability is proposed according to people's demands of daily trips; lastly, the GIS database in the simulation will provide several alternative paths, with which the probabilities of arriving on time and arriving at the optimal time can be calculated for each path, helping UGV select path in accordance with the actual driving tasks. Simulations shown that the proposed path selection method based on GIS database can provide more efficient paths for UGV driving, guiding UGV to complete the driving tasks successfully.
KW - GIS
KW - Path planning
KW - RBF neural network
KW - UGV
UR - https://www.scopus.com/pages/publications/84959353237
U2 - 10.1109/ICCVE.2014.7297642
DO - 10.1109/ICCVE.2014.7297642
M3 - Conference contribution
AN - SCOPUS:84959353237
T3 - 2014 International Conference on Connected Vehicles and Expo, ICCVE 2014 - Proceedings
SP - 705
EP - 710
BT - 2014 International Conference on Connected Vehicles and Expo, ICCVE 2014 - Proceedings
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 3rd International Conference on Connected Vehicles and Expo, ICCVE 2014
Y2 - 3 November 2014 through 7 November 2014
ER -