Multiple-Joint Pedestrian Tracking Using Periodic Models

Feasibility of Using a MEMS Microphone Array for Pedestrian Detection in an Autonomous Emergency Braking System

Pedestrian detection by a car is typically performed using camera, LIDAR, or RADAR-based systems. The first two systems, based on the propagation of light, do not work in foggy or poor visibility environments, and the latter are expensive and the probability associated with their ability to detect people is low. It is necessary to develop systems that are not based on light propagation, with reduced cost and with a high detection probability for pedestrians. This work presents a new sensor that satisfies these three requirements. An active sound system, with a sensor based on a 2D array of MEMS microphones, working in the 14 kHz to 21 kHz band, has been developed. The architecture of the system is based on an FPGA and a multicore processor that allow the system to operate in real time. The algorithms developed are based on a beamformer, range and lane filters, and a CFAR (Constant False Alarm Rate) detector. In this work, tests have been carried out with different people and in different ranges, calculating, in each case and globally, the Detection Probability and the False Alarm Probability of the system. The results obtained verify that the developed system allows the detection and estimation of the position of pedestrians, ensuring that a vehicle travelling at up to 50 km/h can stop and avoid a collision.

Improved Data Association of Hypothesis-Based Trackers Using Fast and Robust Object Initialization

The tracking of Vulnerable Road Users (VRU) is one of the vital tasks of autonomous cars. This includes estimating the positions and velocities of VRUs surrounding a car. To do this, VRU trackers must utilize measurements that are received from sensors. However, even the most accurate VRU trackers are affected by measurement noise, background clutter, and VRUs' interaction and occlusion. Such uncertainties can cause deviations in sensors' data association, thereby leading to dangerous situations and potentially even the failure of a tracker. The initialization of a data association depends on various parameters. This paper proposes steps to reveal the trade-offs between stochastic model parameters to improve data association's accuracy in autonomous cars. The proposed steps can reduce the number of false tracks; besides, it is independent of variations in measurement noise and the number of VRUs. Our initialization can reduce the lag between the first detection and initialization of the VRU trackers. As a proof of concept, the procedure is validated using experiments, simulation data, and the publicly available KITTI dataset. Moreover, we compared our initialization method with the most popular approaches that were found in the literature. The results showed that the tracking precision and accuracy increase to 3.6% with the proposed initialization as compared to the state-of-the-art algorithms in tracking VRU.

A Review of Tracking and Trajectory Prediction Methods for Autonomous Driving

This paper provides a literature review of some of the most important concepts, techniques, and methodologies used within autonomous car systems. Specifically, we focus on two aspects extensively explored in the related literature: tracking, i.e., identifying pedestrians, cars or obstacles from images, observations or sensor data, and prediction, i.e., anticipating the future trajectories and motion of other vehicles in order to facilitate navigating through various traffic conditions. Approaches based on deep neural networks and others, especially stochastic techniques, are reported.