Shortest Path Selection for UAVs using 3-D Coordinates with Collision Avoidance System

Transcription

1 Shortest Path Selection for UAVs using 3-D Coordinates with Collision Avoidance System Shivani Dhiman Department of IT CEC, Landran Mohali, India Mandeep Singh Saini Department of IT CEC, Landran Mohali, India Abstract: The unmanned aerial vehicles are the planes used for the weather application, military applications, area surveillance, etc The unmanned aerial vehicles are cheaper options to the smaller aero-planes which reduce the cost of the operation as well as save the human lives in many cases Also the unmanned aerial vehicles are smaller in size, which makes the undetectable or untraceable on the ordinary radars There are two certain types of the unmanned aerial vehicles (UAVs), where first is automatic UAVs and others are remotely controlled vehicles The path planning and flight management are the real challenges for the automatic UAVs The path planning and flight management or tracing requires a lot of computations which requires the fuel tracking and emergency landing management system for the smoother UAV flights Also the collision arises as one of the major issues for the UAVs The proposed model aims at the path planning and flight management in the case of collision possibility or taking the alternative path due to any hurdle The proposed model has been primarily designed to calculate the point of collision, possibility of collision and alternative path planning (also called bypass routing) The proposed model UAV control and management system has been evaluated for its performance on the basis of various performance parameters The experimental results have proved the efficiency of the proposed model in comparison with the existing models Keywords: UAV routing, Collision avoidance, Alternative path planning, Collision detection I INTRODUCTION An unmanned aerial vehicle (UAV), commonly known as a drone and also referred to as an unpiloted aerial vehicle and a remotely piloted aircraft (RPA) by the International Civil Aviation Organization (ICAO), is an aircraft without a human pilot aboard ICAO classifies unmanned aircraft into two types under Circular 328 AN/190: Autonomous aircraft currently considered unsuitable for regulation due to legal and liability issues Remotely piloted aircraft subject to civil regulation under ICAO and under the relevant national aviation authority There are many different names for these aircraft They are called UAS (unpiloted air system), UAV (unpiloted aerial vehicle), RPAS (remote piloted aircraft systems) and model aircraft It has also become popular to refer to them as drones Their flight is controlled either autonomously by onboard computers or by the remote control of a pilot on the ground or in another vehicle The typical launch and recovery method of an unmanned aircraft is by the function of an automatic system or an external operator on the ground Historically, UAVs were simple remotely piloted aircraft, but autonomous control is increasingly being employed One of the best known and widely used types was the Nazi-German V-1 that flew autonomously powered by a pulse jet Its successor was the Nazi- German V-2, also autonomous but rocket powered, and partly functioning ballistically Unmanned aerial vehicles (UAVs) are actively used for military operations and considered for civilian applications such as environmental monitoring Technological advances in this area for future developments will be to increase the degree of automation of these systems For this we need solutions with acceptable levels of performance to difficult optimization problems, such as variants of the weapon-target assignment problem Often, the problems solved are static combinatorial optimization problems resulting in open loop policies Yet, for most applications of UAVs, involving surveillance and monitoring, we would like to factor into the decision making process the (stochastic) evolution of the environment, which results in even harder stochastic control problems Path planning for small Unmanned Aerial Vehicles (UAVs) is one of the research areas that has received significant attention in the last decade Small UAVs have already been field tested in civilian applications such as wild-fire management, weather and hurricane monitoring and pollutant estimation where the vehicles are used to collect relevant sensor information and transmit the information to the ground (control) stations for further processing Compared to large UAVs, small UAVs are relatively easier to operate and are significantly cheaper Small UAVs can fly at low altitudes and can avoid obstacles or threats at low altitudes more easily Even in military applications, small vehicles are used frequently for intelligence gathering and damage assessment as they are easier to fly and can be hand launched by an individual without any reliance on a runway or a specific type of terrain Even though there are several advantages with using small platforms, they also come with other resource constraints due to their size and limited payload As small UAVs typically have fuel constraints, it may not be possible for an UAV to complete a surveillance mission before refueling at one of the depots For example, consider a typical surveillance mission where a vehicle starts at a depot and is required to visit a set of have been impressive, and it seems now that a major challenge RES Publication 2012 Page 59

2 targets To complete this mission, the vehicle may have to start at the depot, visit a subset of targets and then reach one of the depots for refueling before starting a new path One can reasonably assume that once the UAV reaches a depot, it will be refueled to full capacity before it leaves again for visiting any remaining targets If the goal is to visit each of the given targets at least once, then the UAV may have to repeatedly visit some depots in order to refuel again before visiting all the targets II LITERATURE REVIEW 2014 Manyam, Satyanarayana G et al [1] have proposed the routing algorithm for two Unmanned Aerial Vehicles with communication constraints Contact with UGS is strictly maintained, which allows the UGS act as beacons for relative navigation eliminating the need for dead reckoning This problem is referred to as the Communication Constrained UAV Routing Problem (CCURP) To solve the CCURP, shortest paths between targets are computed by means of a graph transformation Given the shortest paths between targets, two solution methods are presented 2014 Sundar, Kaarthik, and Sivakumar Rathinam [2] have proposed an algorithms for routing an unmanned aerial vehicle in the presence of refueling depots The objective of the problemis to find a path for the UAV such that each target is visited at least once by the vehicle, the fuel constraint is never violated along the path for the UAV, and the total fuel required by the UAV is a minimum 2014 Hernandez-Hernandez, Lucia, Antonios Tsourdos, Hyo-Sang Shin, and Antony Waldock [3] have worked on multi-objective UAV routing method The purpose of the study, presented in this paper, has been to apply a multiobjective graph-based method for plan routes of a simulated UAV taking into account scenarios with danger zones or prohibited areas, characteristic of civil airspace, as well as areas with time and speed restrictions and some negotiable safe corridors 2010 Klein, Daniel J, Johann Schweikl, Jason T Isaacs, and Joao P Hespanha [4] have worked on UAV routing protocols for sparse sensor data exfiltration The problem addressed in this paper is data exfiltration from a collection of sensors that are unable to establish ad-hoc communication due to their widespread deployment, geographical constraints, and power considerations Sensor data is exfiltrated by one or more uninhabited aerial vehicles (UAVs) that act as data mules by visiting each sensor in order to establish a communication link 2008 Ny, Jerome Le, Munther Dahleh, and Eric Feron [5] have done their research on multi-uav dynamic routing with partial observations using restless bandit allocation indices Motivated by the type of missions currently performed by unmanned aerial vehicles, we investigate a discrete dynamic vehicle routing problem with a potentially large number of targets and vehicles III EXPERIMENTAL DESIGN In the proposed model we have focused upon the following parameters, which are used for the working of the movement control and backup path allocation For the hurdle detection or the collision detection, the distance plays the major role in finding the correct location of the hurdle or collision The distance formula is used to compute the gap between the nodes and the hurdle that will be calculated by the following formula: Where are the coordinates of node 1 and are the coordinates of node 2 The displacement can be used to get the movement or distance travelled by the VANET node in the given interval It is the distance calculated of a particular node s location at two different time intervals and is calculated by the following formula: Where is the position of node at time interval t1 and is the position of node at time interval t2 Direction of the movement is important in order to know whether the detect hurdles come in the way of the given VANET node or not The movement of a node from initial location to final location confirms its direction Point of Hurdle (Position) is also an important factor for the proposed model It is the present location of a node which is defined as If the position of node is not moving then it can be termed as a point of hurdle The proposed model is entirely based upon the hazard or hurdle routing for the VANETs The proposed model calculates the node positions and look for the hurdles continuously in the travelling path of the nodes in the cluster The proposed model is aimed at finding and avoiding the hazard situations in the VANET clusters The proposed model is designed to protect the VANETs from the occurring of the collisions due to the hurdles or hazards in the given VANET paths The following is the hurdle or hazard detection algorithm for the VANET cluster: Algorithm 1: Hurdle Detection Algorithm (HDA) 1 The nodes in the cluster scan their locality and objects in the range of their distance sensors deployed for obstacles 2 The nodes are programmed to find the obstacle with the given distance or threshold 3 The programmed nodes actively scan the distance from objects in their way 4 If an object is found on the distance lower than the given threshold a The object is marked as the hurdle b The node stops itself for the moment c All of the following nodes are updated about the hurdle position and situation d The node calculates the alternative path around the hurdle e The node takes the backup path and crosses the hurdle with uninterrupted movement module 5 End IF The nodes in range are updated about the hurdle and take the backup paths according the given alternative path using the hurdle detection algorithm The nodes update their information and follow the calculated instructions by the HDA module RES Publication 2012 Page 60

3 In case, a node is left out of the reach during the transmission range of an RSU (road side unit), the node finds the way to the RSU through the other nodes in its range as well as in the range of the road side unit (RSU) The proposed model is equipped with the hurdle detection algorithm with the unicast updates for the newly joined member in the cluster to update them about the hurdles produced in the given interval The one-hop distance is allowed using the existing model for the connectivity of the unassigned, non-connected and noncovered nodes in the RSU coverage The extended coverage algorithm for the RSU coverage elongation is allowed till the nodes located on the one-hop distance from the member nodes of the given cluster The RSU stores the updates about all of the hurdles produced in the cluster in the given interval length The given interval length is the length of the period for which the RSU will store the information of the hurdles in the unicast table The newly joined nodes are updated about the hurdles found in the given interval by the unicast update propagation model This algorithm is as following: Algorithm 2: UAV Membership and Registration Management 1 When a node found itself not in the reach of any RSU 2 Node sends the query to the nodes in range for their distance from any RSU 3 The nodes reply with their membership with RSU 4 The nodes on the one-hop distance are shortlisted 5 The node with the minimum distance from target node is selected as the next-hop member 6 The node joins the nearest RSU through the next-hop member node 7 The RSU forwards the hurdles updates in the form of unicast update packets 8 The newly joined member will update itself about the hurdle information and will follow the given movement protocol designed for the hurdle avoidance protocol Algorithm 3: Collision Avoidance Algorithm 1 Start the Road side Unit 2 Start the VANET nodes in the cluster 3 Each node shares its coordinate information with all the nodes in cluster Note: The nodes not sending the coordinate information will not become the member of cluster 4 Calculate the following: a Distance b Displacement c Direction d Position e Flexibility for connectivity: The results from the simulation has been obtained in the form of various performance parameters which includes the End-toend delay, Packet Loss, No of Hello Messages (HM) at RSU, No of HM transmitted when 100% vehicles are in hazard zone and Throughput Packet loss increases the retransmissions and hence the network load increases which ultimately gives a rise to end to end delay The proposed work has been entirely based upon the detection and updation of the hurdles in the vehicle paths which are responsible for flooding data and delivering wrong information and the membership of nodes within the VANET cluster For the updates, the vehicles flood the data in the cluster, due to which the network load increases than the normal situations The aim of this research is to lower the data volume during the detection and updation periods in comparison with the existing schemes The proposed model has been designed with the communication efficiency, which controls the data volumes during the proposed work simulation Figure1: End-to-End Delay The transmission delay is the parameter which indicates the total time taken by a packet to travel across the transmission link The delay is added at the maximum rate of 0064 microseconds in the given simulation, which indicates the ideal data transfer across the given simulation The transmission delay indicates the network latency caused due to the load across the path The above graph indicates the comparison with the existing model results which have a maximum delay of 15 So, delay is reduced a lot and network performance is far much better than the existing scheme If Connect to is RSU IV RESULT ANALYSIS Figure 2: Packet Loss RES Publication 2012 Page 61

4 The data loss is the parameters indicator of the loss of the The proposed model records the constant delay after the network convergence period The packet loss has been recorded in comparison with the existing model which has a variable packet loss at a maximum rate of 9 which is very high in comparison packets due to the traffic flow overheads or the bottlenecks at less than 04 %, which is considerably very low The low packet loss indicates the effectiveness of the proposed model Figure 3: Number of HM Received at RSU Number of HM received at RSU is the total number of Hello Messages received at RSUs for making connections with nodes which includes a new connection and connection after any disconnection Thus, high network performance is obtained as the total load for connectivity at RSUs is two messages only Figure 4: Number of HM Transmitted by Vehicles when 100% Vehicles are in Hazard Zone Number of HM transmitted by vehicles to each other or to the RSUs when all the vehicles are in hazard zone and needs timely delivery of messages is reduced from 20 (existing) to 12 (proposed) which increases the reliability and decreases the network load for a better performance in real time scenarios The table given below contains comparison values between the existing and proposed schemes against the parameters of Endto-End Delay, Packet Loss, Number of HM received at RSU, Number of HM transmitted by vehicles when 100% vehicles are in hazard zone: Table 1: Comparison Values of Various Parameters with the Existing Model Sr No End-to-End Delay Packet Loss No of HM Received at RSU No of HM Transmitted by Vehicle when 100% Vehicles are in Hazard Zone Existing Proposed Existing Proposed Existing Proposed Existing Proposed V CONCLUSION AND FUTURE WORK by AODV and EDHRP With the help of EDHRP, network overhead is highly reduced and high reliability is achieved The protocol was implemented and tested using NS235 simulator for transmission range of 250 meters and after the connectivity of the out-of range nodes, the distance covered has been increased up to 500 meters The proposed model can be enhanced for the security of the extended transmission range links through the one-hop nodes Also the more bypass routing methods can be applied to the real time VANET application of the proposed model to create a highly robust and flexible security and movement management model The proposed model has been designed for the automatically driven vehicle based VANETs, which is capable of protecting against the collision or the hurdles produced by the hazards like tree falling, land sliding, or other natural or un-natural obstacles The proposed model is also capable of solving the distance connectivity problems and hurdles produced through the Sybil attacks The proposed model has been evaluated on the basis of various performance parameters The proposed model has performed better than the existing models The experimental results have proved its efficiency which is obtained RES Publication 2012 Page 62

5 REFERENCES [1] Manyam, Satyanarayana G, Sivakumar Rathinam, Swaroop Darbha, David Casbeer, and Phil Chandler "Routing of two Unmanned Aerial Vehicles with communication constraints" In Unmanned Aircraft Systems (ICUAS), 2014 International Conference on, pp IEEE, 2014 [2] Sundar, Kaarthik, and Sivakumar Rathinam "Algorithms for routing an unmanned aerial vehicle in the presence of refueling depots" (2014): 1-8 [3] Hernandez-Hernandez, Lucia, Antonios Tsourdos, Hyo- Sang Shin, and Antony Waldock "Multi-Objective UAV routing" In Unmanned Aircraft Systems (ICUAS), 2014 International Conference on, pp IEEE, 2014 [4] Klein, Daniel J, Johann Schweikl, Jason T Isaacs, and Joao P Hespanha "On UAV routing protocols for sparse sensor data exfiltration" In American Control Conference (ACC), 2010, pp IEEE, 2010 [5] Ny, Jerome Le, Munther Dahleh, and Eric Feron "Multi- UAV dynamic routing with partial observations using restless bandit allocation indices" In American Control Conference, 2008, pp IEEE, 2008 [6] Karaman, S, and E Frazzoli "Linear temporal logic vehicle routing with applications to multi UAV mission planning" International Journal of Robust and Nonlinear Control 21, no 12 (2011): [7] Quintero, Steven AP, Francesco Papi, Daniel J Klein, Luigi Chisci, and João Pedro Hespanha "Optimal UAV coordination for target tracking using dynamic programming" In Decision and Control (CDC), th IEEE Conference on, pp IEEE, 2010 [8] Collins, Gaemus E, James R Riehl, and Philip S Vegdahl "A UAV routing and sensor control optimization algorithm for target search" In Defense and Security Symposium, pp 65610D-65610D International Society for Optics and Photonics, 2007 AUTHOR S BIOGRAPHIES Shivani Dhiman received the BTech Degree in Information Technology from Sri Sai College Of Engineering And Technology Badhani, Pathankot (affiliated to PTU) India in 2012 Currently pursuing MTech in Information Security from CGC, Landran, Mohali (affiliated to PTU) Mandeep Singh Saini received the BTech Degree from Guru Nanak Dev University, Amritsar, Punjab, India And MTech from BBSBEC Fatehgarh Sahib, Punjab, India He is now working as an Assistant Professor in Department of IT in CGC, Landran, Mohali (affiliated to PTU) from 2005 to till date RES Publication 2012 Page 63