project on collision avoidance in static and dynamic environment
TRANSCRIPT
By:Waikhom Pithoijit Singh
Gopal JeeAvinash Kumar
Objective
Motivation
Literature survey
Proposed method
Conclusion and future scope
References
objectiveCollision avoidance in static and dynamic
environment
technique of real time path planning of mobile robot
Motivation Robots send on exploration missionIf robots finds obstacle, sends a signal to
earth stationIn return earth station respond to that signalTime consuming and ineffective in real time
applicationSo this type of inconvenience can be
overcome by the application of collision avoiding robots.
Literature surveyPath planning algorithm for motion planning
by Zidek, k Rigasa, E.Path planning using edge detection method. In this method, an algorithm tries to determine
the position of the vertical edges of the obstacle and then steer the robot around either one of the "visible" edges. The line connecting two visible edges is considered to represent one of the boundaries of the obstacle.
A common drawbacks are poor directionality, frequent misreading, specular reflections.
The Certainty Grid for Obstacle RepresentationIn the certainty grid, the robot's work area is
represented by a two-dimensional array of square elements, denoted as cells. Each cell contains a certainty value (CV) that indicates the measure of confidence that an obstacle exists within the cell area. With the CMU method, CVs are updated by a probability function that takes into account the characteristics of a given sensor
In CMU's applications of this method, the mobile robot remains stationary while it takes a panoramic scan with its 24 ultrasonic sensors
Proposed MethodApplication of ANN for path planning and
obstacle detection.Vector field histogram method using DT for
local path planning.When all the paths are block then we will
used Fuzzy logic.
Artificial Neural Network (ANN)Compound of a large no. of highly
interconnected processing elements (neurons) working in union to solve specific problems.
Loosely modelled on biological neural networkNeuron: processing unit
Fuzzy LogicIt deals with reasoning that is approximate
rather than fixed and exact.Three basic steps involve in fuzzy logic
Fuzzification: changing a real scalar value into a fuzzy value.
Rule Evaluation: an inference is made based on a set of rules.
Defuzzification: the resulting fuzzy output is mapped to a crisp output using the membership functions
WorkingWorking Our target is to avoid collision with the obstacle
in the path.To choose path we are using ANN-DT treeIf all the path are blocked then we will use fuzzy
logic to choose the path.
Future scopeGames and VirtualRobot Motion and NavigationDriverless VehiclesTransportation NetworksHuman Navigation
ReferencesReferences [1] Xiongmin Li, Christine W. Chan , “Application of an enhanced
decision tree learning approach for prediction of petroleum production “ , Engineering Applications of Artificial Intelligence , Elsevier, 23 (2010) 102–109.
[2] Kweku-Muata Osei-Bryson, “Post-pruning in decision tree induction using multiple performance measures”, Computers & Operations Research, Elsevier, 34 (2007) 3331 – 3345.
[3] Hendrik Blockeel *, Luc De Raedt, “Top-down induction of first-order logical decision trees”, Artificial Intelligence, Elsevier, 101 (1998) 285-297.
[4] Max Bramer, “Using J-Pruning to reduce overfitting in classification trees”, Knowledge- Based system, Elsevier, 15(2002) 301-308.
[5] Hussein Almuallim , “An efficient algorithm for optimal pruning of decision trees”, Artificial Intelligence , Elsevier,83 ( 1996) 347-362
[6] J. Borenstein, Member, IEEE and Y. Koren, Senior Member, IEEE, “THE VECTOR FIELD HISTOGRAM -FAST OBSTACLE AVOIDANCE FOR MOBILE ROBOTS », IEEE Journal of Robotics and Automation Vol 7, No 3, June 1991, pp. 278-288.