Vision and navigation for the Carnegie-Mellon navlab
IEEE Transactions on Pattern Analysis and Machine Intelligence - Special Issue on Industrial Machine Vision and Computer Vision Technology:8MPart
Neural network perception for mobile robot guidance
Neural network perception for mobile robot guidance
Selective perception for robot driving
Selective perception for robot driving
Vision-Based Vehicle Guidance
Genetic Algorithms in Search, Optimization and Machine Learning
Genetic Algorithms in Search, Optimization and Machine Learning
Population-Based Incremental Learning: A Method for Integrating Genetic Search Based Function Optimization and Competitive Learning
Removing the Genetics from the Standard Genetic Algorithm
Removing the Genetics from the Standard Genetic Algorithm
Making Driver Modeling Attractive
IEEE Intelligent Systems
An intelligent systems framework for prototyping tactical driving decisions
International Journal of Intelligent Systems Technologies and Applications
A behavioral multi-agent model for road traffic simulation
Engineering Applications of Artificial Intelligence
HSI'09 Proceedings of the 2nd conference on Human System Interactions
A review of the applications of agent technology in traffic and transportation systems
IEEE Transactions on Intelligent Transportation Systems
Modelling tactical driving manoeuvres with GA-INTACT
ICCS'06 Proceedings of the 6th international conference on Computational Science - Volume Part III
Hi-index | 0.00 |
Recent research in automated highway systems has ranged fromlow-level vision-based controllers to high-level route-guidance software.However, there is currently no system for tactical-level reasoning. Such asystem should address tasks such as passing cars, making exits on time, andmerging into a traffic stream. Many previous approaches have attempted tohand construct large rule-based systems which capture the interactionsbetween multiple input sensors, dynamic and potentially conflictingsubgoals, and changing roadway conditions. However, these systems areextremely difficult to design due to the large number of rules, the manualtuning of parameters within the rules, and the complex interactions betweenthe rules. Our approach to this intermediate-level planning is a systemwhich consists of a collection of autonomous agents, each of whichspecializes in a particular aspect of tactical driving. Each agent examinesa subset of the intelligent vehicle‘s sensors and independently recommendsdriving decisions based on their local assessment of the tacticalsituation. This distributed framework allows different reasoning agents tobe implemented using different algorithms.When using a collection of agents to solve a single task, it is vital tocarefully consider the interactions between the agents. Since each reasoningobject contains several internal parameters, manually finding values forthese parameters while accounting for the agents‘ possible interactions is atedious and error-prone task. In our system, these parameters, and thesystem‘s overall dependence on each agent, is automatically tuned using anovel evolutionary optimization strategy, termed Population-BasedIncremental Learning (PBIL).Our system, which employs multiple automatically trained agents, cancompetently drive a vehicle, both in terms of the user-defined evaluationmetric, and as measured by their behavior on several driving situationsculled from real-life experience. In this article, we describe a method formultiple agent integration which is applied to the automated highway systemdomain. However, it also generalizes to many complex robotics tasks wheremultiple interacting modules must simultaneously be configured withoutindividual module feedback.