A Floating-point Extended Kalman Filter Implementation for Autonomous Mobile Robots

Abstract

Localization and Mapping are two of the most important capabilities for autonomous mobile robots and have been receiving considerable attention from the scientific computing community over the last 10 years. One of the most efficient methods to address these problems is based on the use of the Extended Kalman Filter (EKF). The EKF simultaneously estimates a model of the environment (map) and the position of the robot based on odometric and exteroceptive sensor information. As this algorithm demands a considerable amount of computation, it is usually executed on high end PCs coupled to the robot. In this work we present an FPGA-based architecture for the EKF algorithm that is capable of processing twodimensional maps containing up to 1.8 k features at real time (14 Hz), a three-fold improvement over a Pentium M 1.6 GHz, and a 13-fold improvement over an ARM920T 200 MHz. The proposed architecture also consumes only 1.3% of the Pentium and 12.3% of the ARM energy per feature.

A co-processor design to accelerate sequential monocular SLAM EKF process

“…This can be seen from the FPGA implementation by Se et al [16] and Barfoot [17] which utilizes the Scale Invariant Feature Transform (SIFT) feature detection technique to improve SLAM applications. In the estimation stage, Bonato et al [18,19] utilize both Nios II soft core processor and slices inside the Altera Stratix II FPGA to come out with software/hardware solutions to improve the performance of the EKF. The customized hardware design focuses on the matrix multiplication operation which takes most of the processing time.…”

“…The input simplification technique can be further improved if the estimation component itself is improved. This can be seen from Bonato et al [18,19] work which combines both software/hardware FPGA based solution. They proposed a customized hardware matrix multiplication to improve the estimation component of the SLAM process.…”

Localization in vehicular ad hoc networks using data fusion and V2V communication

“…Other services of V2V include 43 transferring the emergency messages, enhancing GPS accuracy, etc. 44 V2I, on the other hand, is applied in toll collection, transferring im- 45 portant localized message related to road-traffic situation as well as 46 providing access for some ITS related services. 47 One of the crucial information attributes in VANETs is location and 48 its accurate calculation is a challenging issue [6].…”

A Hardware Approach for Solving the Robot Localization Problem Using a Sequential EKF

“…addition/subtraction, multiplication, transpose and inverse). This fact suggest the necessity of designing hardware architectures that must take care of memory bandwidth requirements for the FPGA-based EKFs [7], [8], which represents a specific case of the memory-wall problem [9].…”