Author name cluster

Eric Beaudry

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

10 papers

2 author rows

PRL Workshop 2024 Workshop Paper

Conviction-Based Planning for Sparse Reward Reinforcement Learning Problems

Simon Ouellette
Eric Beaudry
Mohamed Bouguessa

Deep reinforcement learning (RL) methods require a large amount of interactions, making them difficult to use in realworld applications. This is especially true when the rewards are sparse since random or semi-random exploration struggles to find them. Learning from Demonstrations (LfD) mitigates this by eliminating the need for random exploration. So far, most LfD solutions have been based on model-free RL approaches that struggle with tasks that require planning. We propose a new algorithm that successfully combines modelbased RL and LfD by leveraging the notion of uncertainty in the transition model during planning. We also introduce the concept of conviction, an uncertainty-to-reward ratio used to decide on optimal actions while planning. Our approach significantly outperforms the relevant baselines in the Minigrid and Sokoban environments.

PDF Details

ECAI Conference 2023 Conference Paper

Cache-Efficient Dynamic Programming MDP Solver

Jaël Champagne Gareau
Guillaume Gosset
Eric Beaudry
Vladimir Makarenkov

Automated planning research often focuses on developing new algorithms to improve the computational performance of planners, but effective implementation can also play a significant role. Hardware features such as memory hierarchy can yield substantial running time improvements when optimized. In this paper, we propose two state-reordering techniques for the Topological Value Iteration (TVI) algorithm. Our first technique organizes states in memory so that those belonging to the same Strongly Connected Component (SCC) are contiguous, while our second technique optimizes state value propagation by reordering states within each SCC. We analyze existing planning algorithms with respect to their cache efficiency and describe domain characteristics which can provide an advantage to each of them. Empirical results show that, in many instances, our new algorithms, called eTVI and eiTVI, run several times faster than traditional VI, TVI, LRTDP and ILAO* techniques.

Details

ICRA Conference 2020 Conference Paper

PARC: A Plan and Activity Recognition Component for Assistive Robots

Jean Massardi
Mathieu Gravel
Eric Beaudry

Mobile robot assistants have many applications, such as helping people in their daily living activities. These robots have to detect and recognize the actions and goals of the humans they are assisting. While there are several wide-spread plan and activity recognition solutions for controlled environments with many built-in sensors, like smart-homes, there is a lack of such systems for mobile robots operating in open settings, such as an apartment. We propose a module for the recognition of activities and goals for daily living by mobile robots, in real time and for complex activities. Our approach recognizes human-object interaction using an RGB-D camera to infer low-level actions which are sent to a goal recognition algorithm. Results show that our approach is both in real time and requires little computational resources, which facilitates its deployment on a mobile and low-cost robotics platform.

Details

ICAPS Conference 2019 Conference Paper

Error-Tolerant Anytime Approach to Plan Recognition Using a Particle Filter

Jean Massardi
Mathieu Gravel
Eric Beaudry

Classical plan recognition approaches require restrictive assumptions and are generally off-line. However, many real-world plan recognition applications must deal with real-time constraints, noisy information, temporal relations in actions, agent preferences, and so on. Many existing approaches have tried to relax assumptions, but none can deal with the above-cited needs. This paper proposes an extension of previous works on plan recognition based on plan tree grammar. Our anytime topdown approach uses a particle filter. This approach manages to give a quick reliable solution to the plan recognition problem while dealing with noisy observations and without reducing the expressiveness of plan libraries. Empirical results on simulated problems show the efficiency of our approach.

Details

TIST Journal 2010 Journal Article

CORALS

Abder Rezak Benaskeur
Froduald Kabanza
Eric Beaudry

Forces involved in modern conflicts may be exposed to a variety of threats, including coordinated raids of advanced ballistic and cruise missiles. To respond to these, a defending force will rely on a set of combat resources. Determining an efficient allocation and coordinated use of these resources, particularly in the case of multiple simultaneous attacks, is a very complex decision-making process in which a huge amount of data must be dealt with under uncertainty and time pressure. This article presents CORALS (COmbat Resource ALlocation Support), a real-time planner developed to support the command team of a naval force defending against multiple simultaneous threats. In response to such multiple threats, CORALS uses a local planner to generate a set of local plans, one for each threat considered apart, and then combines and coordinates them into a single optimized, conflict-free global plan. The coordination is performed through an iterative process of plan merging and conflict detection and resolution, which acts as a plan repair mechanism. Such an incremental plan repair approach also allows adapting previously generated plans to account for dynamic changes in the tactical situation.

Details DOI

IROS Conference 2010 Conference Paper

Motion planning for an omnidirectional robot with steering constraints

Simon Chamberland
Eric Beaudry
Lionel Clavien
Froduald Kabanza
François Michaud
Michel Lauria

Omnidirectional mobile robots, i. e. , robots that can move in any direction without changing their orientation, offer better manoeuvrability in natural environments. Modeling the kinematics of such robots is a challenging problem and different approaches have been investigated. One of the best approaches for a nonholonomic robot is to model the robot's velocity state as the motion around its instantaneous center of rotation (ICR). In this paper, we present a motion planner designed to compute efficient trajectories for such a robot in an environment with obstacles. The action space is modeled in terms of changes of the ICR and the motion around it. Our motion planner is based on a Rapidly-Exploring Random Trees (RRT) algorithm to sample the action space and find a feasible trajectory from an initial configuration to a goal configuration. To generate fluid paths, we introduce an adaptive sampling technique taking into account constraints related to the ICR-based action space.

Details

ICAPS Conference 2010 Conference Paper

Planning for Concurrent Action Executions Under Action Duration Uncertainty Using Dynamically Generated Bayesian Networks

Eric Beaudry
Froduald Kabanza
François Michaud

An interesting class of planning domains, including planning for daily activities of Mars rovers, involves achievement of goals with time constraints and concurrent actions with probabilistic durations. Current probabilistic approaches, which rely on a discrete time model, introduce a blow up in the search state-space when the two factors of action concurrency and action duration uncertainty are combined. Simulation-based and sampling probabilistic planning approaches would cope with this state explosion by avoiding storing all the explored states in memory, but they remain approximate solution approaches. In this paper, we present an alternative approach relying on a continuous time model which avoids the state explosion caused by time stamping in the presence of action concurrency and action duration uncertainty. Time is represented as a continuous random variable. The dependency between state time variables is conveyed by a Bayesian network, which is dynamically generated by a state-based forward-chaining search based on the action descriptions. A generated plan is characterized by a probability of satisfying a goal. The evaluation of this probability is done by making a query the Bayesian network.

Details

ECAI Conference 2010 Conference Paper

Planning with Concurrency under Resources and Time Uncertainty

Eric Beaudry
Froduald Kabanza
François Michaud

Planning with actions concurrency under resources and time uncertainty has been recognized as a challenging and interesting problem. Most current approaches rely on a discrete model to represent resources and time, which contributes to the combinatorial explosion of the search space when dealing with both actions concurrency and resources and time uncertainty. A recent alternative approach uses continuous random variables to represent the uncertainty on time, thus avoiding the state-space explosion caused by the discretization of timestamps. We generalize this approach to consider uncertainty on both resources and time. Our planner is based on a forward chaining search in a state-space where the state representation is characterized by a set of object and numeric state variables. Object state variables are associated with random variables tracking the time at which the state variables' current value has been assigned. The search algorithm dynamically generates a Bayesian network that models the dependency between time and numeric random variables. The planning algorithm queries the Bayesian network to estimate the probability that the resources (numerical state variables) remain in a valid state, the probability of success and the expected cost of the generated plans. Experiments were performed on a transport domain in which we introduced uncertainty on the duration of actions and on the fuel consumption of trucks.

Details

ICAPS Conference 2008 Conference Paper

A Probabilistic Planner for the Combat Power Management Problem

Abder Rezak Benaskeur
Froduald Kabanza
Eric Beaudry
Mathieu Beaudoin

We present a planner for the Combat Power Management (CPM) problem. In response to multiple simultaneous or sequential threats, the planner generates a set of local plans, one for each target considered apart, and then merges them by searching the space of global plans. The proposed plan merging solution serves also as an iterative plan repair process that resolves negative interferences (subaddivity) and exploits synergistic effects (superaddivity) among activities. The planner was developed as a component of a naval Command and Control system to support the warship command team in defending against Anti-Ship Missile threats. The planner is particularly suited for domains characterized by durative concurrent actions, with both superadditive and subadditive interactions, and probabilistic effects. It is implemented in a generic way, allowing it to address other application domains. We discuss the CPM application domain, describe the planner, and present experimental results.

Details

IROS Conference 2008 Conference Paper

Reactive planning as a motivational source in a behavior-based architecture

Eric Beaudry
Dominic Létourneau
Froduald Kabanza
François Michaud

Behavior-based architectures use behaviors as building blocks for decision-making and action execution processes. Behaviors are distributed and evaluated in parallel for the control of the robot, taking real-time inputs from sensory data and sending real-time commands to effectors. No centralized components exist in these architectures, each module carrying out its own strategy independently, making an overall behavior emerge from the interaction between the concurrently executed modules and the environment. In this paper, we discuss the use of a reactive hierarchical task network (HTN) planner in a behavior-based robot architecture. The planner in this architecture is not a central component on which everything else relies on, but acts as one of the motivational modules recommending tasks to be executed and influencing the selection and configuration of behaviors. The planning module allows the behavior-based architecture to deal with tasks with priorities, flexible time constraints and on-line planning using a simple but very effective reactive planning strategy. We demonstrate our approach in the context of making a robot attend a conference.

Details