Weighted Node Mapping and Localisation on a Pixel Processor Array

This paper implements and demonstrates visual route mapping and localisation upon a Pixel Processor Array (PPA). The PPA sensor comprises of an array of Processing Elements (PEs), each of which can capture and process visual information directly. This provides significant parallel processing power allowing novel ways in which information can be processed on-sensor. Our method predicts the correct node within a topological map generated from an image sequence by measuring image similarities, spatial coherence, and exploiting the parallel nature of the PPA. Our implementation runs at +300Hz on large public datasets with +2K locations requiring 2. 5W at 500 GOPS/W. We compare vs traditionally implemented methods demonstrating better F-1 performance even on simulation. As far as we are aware, we present the first on-sensor mapping and localisation system running entirely on-sensor.

Authors

• Hector Castillo-Elizalde
• Yanan Liu
• Laurie Bose
• Walterio W. Mayol-Cuevas

Keywords

• Visualization
• Spatial coherence
• Prototypes
• Parallel processing
• Robot sensing systems
• Hardware
• Computational efficiency
• Node Mapping
• Topological Map
• Large Public Datasets
• Similarity Measure
• Imaging Methods
• Input Image
• F1 Score
• Parallelization
• Simulation Environment
• Local Method
• Motion Model
• Current Image
• Low-resolution Images
• Particle Filter
• Image Database
• Hamming Distance
• Image Descriptors
• Binary Method
• Node Weights
• Sum Of Absolute Differences
• Real Hardware
• Temporal Weights
• Focal Plane
• Time Step