Sensing and Sensor Fusion for the 2005 Desert Buckeyes DARPA Grand Challenge Offroad Autonomous Vehicle

Abstract

This paper describes the sensor suite and the sensor fusion algorithms used for external environment sensing in the Ohio State University Desert Buckeyes 2005 DARPA Grand Challenge autonomous off-road vehicle entry. The external environment sensor suite included laser range£nders, a stereo vision system, a radar system, and ultrasonic range£nders. The positioning and orientation sensor suite included a precision GPS receiver, a vertical gyroscope, a digital compass, and dead reckoning wheel speed sensors. Sensor fusion was accomplished on a cell-based map using multiple con£dence measures and estimated object height. Issues arising from terrain and the estimation of a ground pro£le ahead of the vehicle are also discussed.

Multi-layer occupancy grid mapping for autonomous vehicles navigation

“…But with the increased number of sensors we experienced also a growth in the complexity of the task of map creation. From the results of the first years of the century different approaches to the problem have been proposed, in particular many researches are focused on achieving the best possible results on a specific task like parking [13], driving in off road scenarios [17], or using only 978-8-8872-3743-6 ©2019 AEIT specific sensors like radar [14]. Nevertheless, during past years some more complex and complete systems has been proposed, [15], [16].…”

Sensing, Control, and System Integration for Autonomous Vehicles: A Series of Challenges

“…For AHS the reader is referred to our Demo'97 vehicle [1], shown in Fig. 1, and for offroad driving our DARPA Grand Challenge 2004 vehicle TerraMax [2] and 2005 vehicle ION, the Intelligent Offroad Navigator [3], [4], shown in Figs. 2a and b.…”

“…This means that the map doesn't rotate with the vehicle, but it does translate. The sensor fusion algorithm implemented on OSU's 2005 DARPA Grand Challenge vehicle used such a grid occupancy approach [4].…”

Discrete-State Encoding in Hybrid-State Systems for Intelligent Vehicle Control and Estimation

“…The coupling between these systems of different domains has successfully been modeled as hybrid-state systems (HSSs), for a wide variety of cyberphysical systems, including autonomous vehicle applications [1], [2] and estimation/prediction of driver/vehicle interactions [3] at The Ohio State University Control and Intelligent Transportation Research Laboratory.…”