Tirthankar Bandyopadhyay

Intention-Aware Motion Planning

As robots venture into new application domains as autonomous vehicles on the road or as domestic helpers at home, they must recognize human intentions and behaviors in order to operate effectively. This paper investigates a new class of motion planning problems with uncertainty in human intention. We propose a method for constructing a practical model by assuming a finite set of unknown intentions. We first construct a motion model for each intention in the set and then combine these models together into a single Mixed Observability Markov Decision Process (MOMDP), which is a structured variant of the more common Partially Observable Markov Decision Process (POMDP). By leveraging the latest advances in POMDP/MOMDP approximation algorithms, we can construct and solve moderately complex models for interesting robotic tasks. Experiments in simulation and with an autonomous vehicle show that the proposed method outperforms common alternatives because of its ability in recognizing intentions and using the information effectively for decision making.

Synthetic 2D LIDAR for precise vehicle localization in 3D urban environment

This paper presents a precise localization algorithm for vehicles in 3D urban environment with only one 2D LIDAR and odometry information. A novel idea of synthetic 2D LIDAR is proposed to solve the localization problem on a virtual 2D plane. A Monte Carlo Localization scheme is adopted for vehicle position estimation, based on synthetic LIDAR measurements and odometry information. The accuracy and robustness of the proposed algorithm are demonstrated by performing real time localization in a 1.5 km driving test around the NUS campus area.

Curb-intersection feature based Monte Carlo Localization on urban roads

One of the most prominent features on an urban road is the curb, which defines the boundary of a road surface. An intersection is a junction of two or more roads, appearing where no curb exists. The combination of curb and intersection features and their idiosyncrasies carry significant information about the urban road network that can be exploited to improve a vehicles localization. This paper introduces a Monte Carlo Localization (MCL) method using the curb-intersection features on urban roads. We propose a novel idea of “Virtual LIDAR” to get the measurement models for these features. Under the MCL framework, above road observation is fused with odometry information, which is able to yield precise localization. We implement the system using a single tilted 2D LIDAR on our autonomous test bed and show robust performance in the presence of occlusion from other vehicles and pedestrians.

Autonomous personal vehicle for the first- and last-mile transportation services

This paper describes an autonomous vehicle testbed that aims at providing the first- and last- mile transportation services. The vehicle mainly operates in a crowded urban environment whose features can be extracted a priori. To ensure that the system is economically feasible, we take a minimalistic approach and exploit prior knowledge of the environment and the availability of the existing infrastructure such as cellular networks and traffic cameras. We present three main components of the system: pedestrian detection, localization (even in the presence of tall buildings) and navigation. The performance of each component is evaluated. Finally, we describe the role of the existing infrastructural sensors and show the improved performance of the system when they are utilized.

Intention-Aware Pedestrian Avoidance

A critical component of autonomous driving in urban environment is the vehicle’s ability to interact safely and intelligently with the human drivers and on-road pedestrians. This requires identifying the human intentions in real time based on a limited observation history and reacting accordingly. In the context of pedestrian avoidance, traditional approaches like proximity based reactive avoidance, or taking the most likely behavior of the pedestrian into account, often fail to generate a safe and successful avoidance strategy. This is mainly because they fail to take into account the human intention and the inherent uncertainty resulting in identifying such intentions from direct observations.

Motion Planning for 3-D Target Tracking among Obstacles

The goal of target tracking is to compute motion strategies for a robot equipped with visual sensors, so that it can effectively track a moving target despite obstruction by obstacles. It is an important problem with many applications in robotics. Existing work focuses mostly on the 2-D version of the problem, partly due to the complexity of dealing with 3-D visibility relationships. This paper proposes an online algorithm for 3-D target tracking among obstacles, using only local geometric information available to a robot’s visual sensors. Key to this new algorithm is the definition and efficient computation of a risk function, which tries to capture a target’s ability in escaping from the robot sensors’ visibility region in both short and long terms. The robot then moves to minimize this risk function locally in a greedy fashion. In the absence of occlusion by obstacles, the standard tracking algorithm based on visual servo control can be considered a special case of our algorithm. Experiments show that the new algorithm generated interesting tracking behaviors in three dimensions and performed substantially better than visual servo control in simulation.

A general framework for road marking detection and analysis

Road markings are paintings on road surface to provide traffic guidance information for vehicles and pedestrians. In this paper, we propose a general framework for road marking detection and analysis, which is able to support various types of markings. Marking contours of different types are extracted indiscriminately from a image processing procedure, and sent to respective modules for independent classification and analysis. Four common types of markings are studied as examples in this paper, including lanes, arrows, zebra-crossings, and words. Our proposed method is tested through experiments, and shows good performance.

A Spatial-Temporal Approach for Moving Object Recognition with 2D LIDAR

Moving object recognition is one of the most fundamental functions for autonomous vehicles, which occupy an environment shared by other dynamic agents. We propose a spatial-temporal (ST) approach for moving object recognition using a 2D LIDAR. Our experiments show reliable performance. The contributions of this paper include: (i) the design of ST features for accumulated 2D LIDAR data; (ii) a real-time implementation for moving object recognition using the ST features.

Mapping with synthetic 2D LIDAR in 3D urban environment

In this paper, we report a fully automated detailed mapping of a challenging urban environment using single LIDAR. To improve scan matching, extended correlative scan matcher is proposed. Also, a Monte Carlo loop closure detection is implemented to perform place recognition efficiently. Automatic recovery of the pose graph map in the presence of false place recognition is realized through a heuristic based loop closure rejection. This mapping framework is evaluated through experiments on the real world dataset obtained from NUS campus environment.

Local Voronoi Decomposition for multi-agent task allocation

We propose a Local Voronoi Decomposition (LVD) Algorithm which is able to perform a robust and online task allocation for multiple agents based purely on local information. Because only local information is required in determining each agents Voronoi region, each agent can then make its decision in a distributive fashion based on its allocated Voronoi region. These Voronoi regions eliminates the occurrence of agents executing instantaneous overlapping tasks. As our method does not require a pre-processing of the map, it is also able to work well in a dynamically changing map with changing number of agents. We will show our proof of concept in the problem of exploration in an unknown environment. In our experimental evaluation, we show that our method significantly outperforms the competing algorithms: Ants Algorithm and the Brick&Mortar Algorithm. Our results also show that our method is near the theoretical best solution.