Inna Sharf
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Research Interests
Inna Sharf has a total of 52 co-authors. Here are the top co-authors by number of shared publications:
- Catherine Massé
Department of Mechanical Engineering, McGill University, Montreal, Canada
4 shared publications - David Zarrouk
Department of Mechanical Engineering, Ben Gurion University of the Negev, Beersheba, Israel
3 shared publications - Moshe Shoham
Technion – Israel Institute of Technology
3 shared publications - Eleonora Marina Botta
University at Buffalo
2 shared publications Florent Deleflie
2 shared publications- Jiazhi Song
Department of Mechanical Engineering, McGill University, Montreal, QC, Canada
2 shared publications
Latest Publications
- 3 March 2026
- Published by Springer Nature in The Journal of the Astronautical Sciences
- Vol. 73 (2) , 27
Resonant Motion of Echo I: A Case Study of SRP and $J_2$ Coupling- 3 March 2026
- Published by Springer Nature in The Journal of the Astronautical Sciences
- Vol. 73 (2) , 27
Resonant Motion of Echo I: A Case Study of SRP and $J_2$ Coupling- Open Access
- 24 October 2025
- Published by Elsevier in Aerospace Science and Technology
- Vol. 168
Momentum transfer by impact: A step towards grasping of non-cooperative space debris
- 24 October 2025
- Published by Elsevier in Aerospace Science and Technology
- Vol. 168
Open AccessMomentum transfer by impact: A step towards grasping of non-cooperative space debris
- Open Access
- 18 August 2025
- Published by Springer Nature in Research Square
Resonant motion of Echo I: a case study of SRP and J2 coupling
In this paper, we explore the resonant behaviour exhibited by the historical NASA’s Echo I satellite. It is well known that the coupling between the SRP and the geopotential (second zonal harmonic 𝑱2) perturbations leads to a resonance effect in eccentricity under certain orbital conditions. Here, we revisit the orbital motion of Echo I in light of the recently published phase space theory for spherical spacecraft subject to the SRP-𝑱2 resonance. We first demonstrate that our modelling is in very good agreement with the recorded Two Line Element (TLE) data of the spacecraft. We then derive the first ( 𝒋 = 1) resonance model associated with the motion of Echo I by using the Hamiltonian formulation of the satellite dynamics, which we use to interpret its observed behaviour. Finally, we extrapolate the resonance model of Echo I for two imagined extreme scenarios: the motion near a stable equilibrium, and the motion under the strongest resonance possible. The results provide a new understanding of the orbital motion of Echo I and offer new insights into the SRP-𝑱2 resonance phenomenon.
- 18 August 2025
- Published by Springer Nature in Research Square
Open AccessResonant motion of Echo I: a case study of SRP and J2 coupling
In this paper, we explore the resonant behaviour exhibited by the historical NASA’s Echo I satellite. It is well known that the coupling between the SRP and the geopotential (second zonal harmonic 𝑱2) perturbations leads to a resonance effect in eccentricity under certain orbital conditions. Here, we revisit the orbital motion of Echo I in light of the recently published phase space theory for spherical spacecraft subject to the SRP-𝑱2 resonance. We first demonstrate that our modelling is in very good agreement with the recorded Two Line Element (TLE) data of the spacecraft. We then derive the first ( 𝒋 = 1) resonance model associated with the motion of Echo I by using the Hamiltonian formulation of the satellite dynamics, which we use to interpret its observed behaviour. Finally, we extrapolate the resonance model of Echo I for two imagined extreme scenarios: the motion near a stable equilibrium, and the motion under the strongest resonance possible. The results provide a new understanding of the orbital motion of Echo I and offer new insights into the SRP-𝑱2 resonance phenomenon.
- 11 July 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Robotics
- Vol. 41 (15523098) , 4574-4587
Baseline Policy Adapting and Abstraction of Shared Autonomy For High-Level Robot Operations
This paper presents a novel shared autonomy and baseline policy adapting framework for human-robot interactions in high-level context-aware robotic tasks. With a unique methodology that leverages hierarchies in decision-making as well as variational analysis of human policy, we propose a mathematical model of shared autonomy policy. The framework aims at interpretable high-level decision-making for efficient robot operation with human in the loop. We modeled the decision-making process using hierarchical Markov decision processes (MDPs) in an algorithm we called policy adapting, where the autonomous system policy is adapted, and hence, shaped by incorporating design variables contextual to the robot, human, task, and pre-training. By integrating deep reinforcement learning within a multi-agent hierarchical context, we present an end-to-end algorithm to train a baseline policy designed for shared autonomy. We showcase the effectiveness of our framework, and particularly the interplay between different design elements and human's skill level, in a pilot study with a human user in a simulated sequence of high-level pick-and-place tasks. The proposed framework advances the state-of-the-art in shared autonomy for robotic tasks, but can also be applied to other domains of autonomous operation.
- 11 July 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Robotics
- Vol. 41 (15523098) , 4574-4587
Baseline Policy Adapting and Abstraction of Shared Autonomy For High-Level Robot Operations
This paper presents a novel shared autonomy and baseline policy adapting framework for human-robot interactions in high-level context-aware robotic tasks. With a unique methodology that leverages hierarchies in decision-making as well as variational analysis of human policy, we propose a mathematical model of shared autonomy policy. The framework aims at interpretable high-level decision-making for efficient robot operation with human in the loop. We modeled the decision-making process using hierarchical Markov decision processes (MDPs) in an algorithm we called policy adapting, where the autonomous system policy is adapted, and hence, shaped by incorporating design variables contextual to the robot, human, task, and pre-training. By integrating deep reinforcement learning within a multi-agent hierarchical context, we present an end-to-end algorithm to train a baseline policy designed for shared autonomy. We showcase the effectiveness of our framework, and particularly the interplay between different design elements and human's skill level, in a pilot study with a human user in a simulated sequence of high-level pick-and-place tasks. The proposed framework advances the state-of-the-art in shared autonomy for robotic tasks, but can also be applied to other domains of autonomous operation.
- 16 December 2024
- Published by Springer Nature in Nonlinear Dynamics
- Vol. 113 (8) , 8115-8137
Phase space analysis and validation of resonant dynamics of rotating plate-like spacecraft
- 16 December 2024
- Published by Springer Nature in Nonlinear Dynamics
- Vol. 113 (8) , 8115-8137
Phase space analysis and validation of resonant dynamics of rotating plate-like spacecraft
- 5 November 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Automation Science and Engineering
- Vol. 22 (15455955) , 8439-8452
Motion Planners for Path or Waypoint Following and End-Effector Sway Damping With Dynamic Programming
We propose two novel motion planners for a robotic manipulator with a passive end-effector that is free to sway during and after the robot’s motion. The planners utilize Dynamic Programming to generate trajectories that damp the end-effector’s residual sway while ensuring that the boom tip—the point to which the end-effector is attached—follows a collision-free path or time-dependent waypoints. Our use case is a crane of a forwarder machine, a log-loading machine in the forestry industry, with a passive grapple. In the cluttered forest environment, accurate path following and grapple sway damping are critical to increase the operation efficiency and avoid harming the machine and environment. The results of the simulation in a high-fidelity multibody-dynamics simulator showcase the effectiveness of our methodology in achieving exact path following or timed waypoints following and the residual sway damping. In particular, the average of the maximum residual sway is only $1.9^\circ$ , showing an average reduction of $75\%$ , as compared to fifth, sixth, and tenth order polynomial trajectories, in six test cases, including common paths used by operators to pick and place logs. Monte-Carlo simulations also showed that our planners have very good robustness against payload mass uncertainty. Other merits of our Dynamic Programming trajectories are that they are smooth, computationally inexpensive, and result in reduced residual sway even for nonzero initial sway conditions. Moreover, the generality of our methodology opens a new way to design anti-sway motion planners for construction cranes or quadrotors with a slung payload, in addition to serial manipulators with passive end-effectors. Note to Practitioners —This work was motivated by the problems arising in the operation of log-loading cranes in the forestry industry: the problems of the end-effector’s large sway during crane reconfiguration and the collision between the crane and obstacles, which are detrimental to the efficiency of the operation. Similar issues arise, for example, in construction cranes transporting large hanging objects. We propose a novel methodology to address both problems by generating smooth and computationally inexpensive trajectories for the crane joint motion. The approach begins with the model of the sway motion and the definition of the collision-free path. Then, our Dynamic Programming algorithm generates anti-sway trajectories that satisfy the joint constraints. The results in a high-fidelity simulator show that our motion planners lead to precise path following and significant sway damping, and also confirm its superiority compared to polynomial trajectories, commonly used in industries. Monte-Carlo simulation also confirms our planners’ robustness. Our methodology is also applicable to other dynamic systems with freely hanging objects, such as multi-degree-of-freedom robotic manipulators, construction cranes, and quadrotors carrying a slung payload. A possible limitation is that the methodology is model-based and necessitates finding the sway dynamics model and estimating payload properties.
- 5 November 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Automation Science and Engineering
- Vol. 22 (15455955) , 8439-8452
Motion Planners for Path or Waypoint Following and End-Effector Sway Damping With Dynamic Programming
We propose two novel motion planners for a robotic manipulator with a passive end-effector that is free to sway during and after the robot’s motion. The planners utilize Dynamic Programming to generate trajectories that damp the end-effector’s residual sway while ensuring that the boom tip—the point to which the end-effector is attached—follows a collision-free path or time-dependent waypoints. Our use case is a crane of a forwarder machine, a log-loading machine in the forestry industry, with a passive grapple. In the cluttered forest environment, accurate path following and grapple sway damping are critical to increase the operation efficiency and avoid harming the machine and environment. The results of the simulation in a high-fidelity multibody-dynamics simulator showcase the effectiveness of our methodology in achieving exact path following or timed waypoints following and the residual sway damping. In particular, the average of the maximum residual sway is only $1.9^\circ$ , showing an average reduction of $75\%$ , as compared to fifth, sixth, and tenth order polynomial trajectories, in six test cases, including common paths used by operators to pick and place logs. Monte-Carlo simulations also showed that our planners have very good robustness against payload mass uncertainty. Other merits of our Dynamic Programming trajectories are that they are smooth, computationally inexpensive, and result in reduced residual sway even for nonzero initial sway conditions. Moreover, the generality of our methodology opens a new way to design anti-sway motion planners for construction cranes or quadrotors with a slung payload, in addition to serial manipulators with passive end-effectors. Note to Practitioners —This work was motivated by the problems arising in the operation of log-loading cranes in the forestry industry: the problems of the end-effector’s large sway during crane reconfiguration and the collision between the crane and obstacles, which are detrimental to the efficiency of the operation. Similar issues arise, for example, in construction cranes transporting large hanging objects. We propose a novel methodology to address both problems by generating smooth and computationally inexpensive trajectories for the crane joint motion. The approach begins with the model of the sway motion and the definition of the collision-free path. Then, our Dynamic Programming algorithm generates anti-sway trajectories that satisfy the joint constraints. The results in a high-fidelity simulator show that our motion planners lead to precise path following and significant sway damping, and also confirm its superiority compared to polynomial trajectories, commonly used in industries. Monte-Carlo simulation also confirms our planners’ robustness. Our methodology is also applicable to other dynamic systems with freely hanging objects, such as multi-degree-of-freedom robotic manipulators, construction cranes, and quadrotors carrying a slung payload. A possible limitation is that the methodology is model-based and necessitates finding the sway dynamics model and estimating payload properties.
- Open Access
- 24 July 2024
- Published by Elsevier in Acta Astronautica
- Vol. 223, 666-684
Model-based controllers for CubeSat ORU installation: A comparative study
- 24 July 2024
- Published by Elsevier in Acta Astronautica
- Vol. 223, 666-684
Open AccessModel-based controllers for CubeSat ORU installation: A comparative study
- 4 December 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE/ASME Transactions on Mechatronics
- Vol. 29 (4) , 2579-2589
Chance-Constrained Rollover-Free Manipulation Planning With Uncertain Payload Mass
This article presents a chance-constrained rollover-free manipulation planning method for robotic arms under payload mass uncertainty. The corresponding motion planning problem is stated as a chance-constrained nonlinear optimal control problem (NOCP) subject to kinematics and rollover stability constraints. The latter takes the form of a chance constraint that ensures a certain probability of the robot maintaining dynamic rollover stability in the presence of payload mass uncertainty. To achieve efficient solutions to the NOCP, a novel geometric bound for the stability region is derived. The novel bound is then utilized to modify the rollover-stability constraint. To showcase its benefit, comparisons between the proposed bound of probabilistic rollover-stability measure and the naive noise model are provided through statistical analysis. The formulation's practicality is demonstrated through experiments with a Kinova Jaco2 arm mounted on a free-to-rollover platform. Results demonstrate greater robustness of the robot's motion plan to mass uncertainty and computational efficiency of the trajectory generation.
- 4 December 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE/ASME Transactions on Mechatronics
- Vol. 29 (4) , 2579-2589
Chance-Constrained Rollover-Free Manipulation Planning With Uncertain Payload Mass
This article presents a chance-constrained rollover-free manipulation planning method for robotic arms under payload mass uncertainty. The corresponding motion planning problem is stated as a chance-constrained nonlinear optimal control problem (NOCP) subject to kinematics and rollover stability constraints. The latter takes the form of a chance constraint that ensures a certain probability of the robot maintaining dynamic rollover stability in the presence of payload mass uncertainty. To achieve efficient solutions to the NOCP, a novel geometric bound for the stability region is derived. The novel bound is then utilized to modify the rollover-stability constraint. To showcase its benefit, comparisons between the proposed bound of probabilistic rollover-stability measure and the naive noise model are provided through statistical analysis. The formulation's practicality is demonstrated through experiments with a Kinova Jaco2 arm mounted on a free-to-rollover platform. Results demonstrate greater robustness of the robot's motion plan to mass uncertainty and computational efficiency of the trajectory generation.
- 7 November 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Transportation Systems
- Vol. 25 (5) , 3898-3910
Chance-Constrained Planning for Dynamically Stable Motion of Reconfigurable Vehicles
In this paper, a computationally efficient chance-constrained rollover-free motion planning method is presented. Specifically, the method is developed to plan motions for reconfigurable vehicles with the knowledge of a 3-D terrain model that has limited accuracy. The overall motion planning problem is formulated as a nonlinear optimal control problem (NOCP) that employs a constraint in the form of a bound on the probability of rollover under terrain-induced vehicle orientation uncertainty. To increase the computational efficiency of the NOCP with nonlinear chance constraint, a geometric interpretation of the chance constraint is derived based on the characteristics of SO(3), the 3-D rotation group. Monte Carlo simulations are provided to demonstrate the usefulness of the geometric interpretation through comparisons with other methods. Experimental data gathered from driving a mobile robot through real forests are also used to validate the proposed model. Finally, path and trajectory generation results obtained with the proposed planning method for a feller-buncher machine traversing through uncertain 3-D terrain are presented to showcase the method’s overall performance and efficiency.
- 7 November 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Transportation Systems
- Vol. 25 (5) , 3898-3910
Chance-Constrained Planning for Dynamically Stable Motion of Reconfigurable Vehicles
In this paper, a computationally efficient chance-constrained rollover-free motion planning method is presented. Specifically, the method is developed to plan motions for reconfigurable vehicles with the knowledge of a 3-D terrain model that has limited accuracy. The overall motion planning problem is formulated as a nonlinear optimal control problem (NOCP) that employs a constraint in the form of a bound on the probability of rollover under terrain-induced vehicle orientation uncertainty. To increase the computational efficiency of the NOCP with nonlinear chance constraint, a geometric interpretation of the chance constraint is derived based on the characteristics of SO(3), the 3-D rotation group. Monte Carlo simulations are provided to demonstrate the usefulness of the geometric interpretation through comparisons with other methods. Experimental data gathered from driving a mobile robot through real forests are also used to validate the proposed model. Finally, path and trajectory generation results obtained with the proposed planning method for a feller-buncher machine traversing through uncertain 3-D terrain are presented to showcase the method’s overall performance and efficiency.
- Open Access
Hierarchical Planning and Policy Shaping Shared Autonomy for Articulated Robots
In this work, we propose a novel shared autonomy framework to operate articulated robots. We provide strategies to design both the task-oriented hierarchical planning and policy shaping algorithms for efficient human-robot interactions in context-aware operation of articulated robots. Our framework for interplay between the human and the autonomy, as the participating agents in the system, is particularly influenced by the ideas from multi-agent systems, game theory, and theory of mind for a sliding level of autonomy. We formulate the sequential hierarchical human-in-the-loop decision making process by extending MDPs and Options framework to shared autonomy, and make use of deep RL techniques to train an uncertainty-aware shared autonomy policy. To fine-tune the formulation to a human, we use history of the system states, human actions, and their error with respect to a surrogate optimal model to encode human's internal state embeddings, beyond the designed values, by using conditional VAEs. We showcase the effectiveness of our formulation for different human skill levels and degrees of cooperativeness by using a case study of a feller-buncher machine in the challenging tasks of timber harvesting. Our framework is successful in providing a sliding level of autonomy from fully autonomous to fully manual, and is particularly successful in handling a noisy non-cooperative human agent in the loop. The proposed framework advances the state-of-the-art in shared autonomy for operating articulated robots, but can also be applied to other domains where autonomous operation is the ultimate goal.
Open AccessHierarchical Planning and Policy Shaping Shared Autonomy for Articulated Robots
In this work, we propose a novel shared autonomy framework to operate articulated robots. We provide strategies to design both the task-oriented hierarchical planning and policy shaping algorithms for efficient human-robot interactions in context-aware operation of articulated robots. Our framework for interplay between the human and the autonomy, as the participating agents in the system, is particularly influenced by the ideas from multi-agent systems, game theory, and theory of mind for a sliding level of autonomy. We formulate the sequential hierarchical human-in-the-loop decision making process by extending MDPs and Options framework to shared autonomy, and make use of deep RL techniques to train an uncertainty-aware shared autonomy policy. To fine-tune the formulation to a human, we use history of the system states, human actions, and their error with respect to a surrogate optimal model to encode human's internal state embeddings, beyond the designed values, by using conditional VAEs. We showcase the effectiveness of our formulation for different human skill levels and degrees of cooperativeness by using a case study of a feller-buncher machine in the challenging tasks of timber harvesting. Our framework is successful in providing a sliding level of autonomy from fully autonomous to fully manual, and is particularly successful in handling a noisy non-cooperative human agent in the loop. The proposed framework advances the state-of-the-art in shared autonomy for operating articulated robots, but can also be applied to other domains where autonomous operation is the ultimate goal.