When: Wednesday, April 6th at 15:00
Where: Zoom
Abstract: Classic state estimation tools (e.g., determining position/velocity of a robot from noisy sensor data) have been in use since the 1960s, perhaps the most famous technique being the Kalman filter. For difficult-to-model nonlinear systems with rich sensing (e.g., almost any real-world robot), clever adaptations are needed to the classic tools. In this talk, I will first briefly summarize an idea that has become standard practice in our group over the last several years: continuous-time trajectory estimation (and its connection to sparse Gaussian process regression). I will then discuss two new frameworks we have been pursuing lately: exactly sparse Gaussian variationally inference (ESGVI) and Koopman state estimation (KoopSE). ESGVI seeks to minimize the Kullback-Leibler divergence between a Gaussian state estimate and the full Bayesian posterior; however, the framework also easily allows for parameter learning through Expectation Maximization and we’ve used this to learn simple parameters such as constant system matrices and covariances, but also to model rich sensors using Deep Neural Networks and learn the weights from data. KoopSE takes a different approach by lifting a nonlinear system into a high-dimensional Reproducing Kernel Hilbert Space where we can treat it as linear and apply classic estimation tools; it also allows for the system to be learned from training data quite efficiently. I will give simple intuitive explanations of the mathematics and show some examples of things working in practice.
Work towards PhD degree under the supervision of Prof. Amir Degani
When: 14.4.2022 at 11:00
Where: Zoom
The seminar will be held in hybrid format: Water institute auditorium
Abstract: In recent years, there has been considerable advancement in the fields of planning and control of legged robots, but legged robots are still extremely limited. Most research effort has been directed toward bipedal walking and quadrupedal walking and running, and not so much at bipedal running. Dynamic-legged locomotion is difficult to plan and execute, it requires robust control strategies and fast planning algorithms. In this work, we describe a strategy for traversing uneven complex terrain without the need to perform explicit foot placement. This will be achieved using a funnel library approach in combination with an online receding horizon planner. Thus, enabling the robot to take full advantage of its dynamics and the potential reaction forces presented by the environment. We demonstrate, in simulation, multiple scenarios, from traversing a low-friction obstacle, all the way to inclined surfaces and chute climbing with intermittent low-friction patches. We present an efficient approach for generating transitions from running to climbing without the need for a long planning horizon. We investigate the SLIP, ASLIP, and the actuated SLIP models and several in-stride control approaches.
When: 23.3.2022 at 15:00
Where: zoom
Abstract: In 2016, Schlumberger started to deploy automated drilling, using plan-based control, across multiple rigs in several basins. Five years later, the technology has drilled more than half-a-million feet under fully autonomous control, with an estimated 17% efficiency improvement over human operations. A significant challenge in this deployment has been the range of rigs on which the system has been deployed, each with different degrees of access to control systems, different sensors and all of them in different geophysical settings. In this talk I will outline the planning technology and the benefits it has brought in the achievement of this success, the approaches we have adopted to interfacing between planning and execution, the scope of some of the other uses we have found for the same approach and, depending on time, some of the planning-related problems we are currently investigating.
Work towards MSc degree under the supervision of Dr. Oren Salzman
When: 7.3.2022 at 9:30
Where: zoom
Abstract: In Lifelong Multi-Agent Path Finding (L-MAPF) a team of agents performs a stream of tasks consisting of multiple locations to be visited by the agents on a shared graph while avoiding collisions with one another. L-MAPF is typically tackled by partitioning it into multiple consecutive, and hence similar, “one-shot” MAPF queries with a single task assigned to each agent, as in the Rolling-Horizon Collision Resolution (RHCR) algorithm. Therefore, a solution to one query informs the next query, which leads to similarity with respect to the agents’ start and goal positions, and how collisions between the agents need to be resolved from one query to the next. Thus, experience from solving one MAPF query can potentially be used to speedup solving the next one and reduce runtime in L-MAPF overall. Despite this intuition, current L-MAPF planners solve consecutive MAPF queries from scratch. In this paper, we introduce a new RHCR-inspired approach called exRHCR, which exploits experience in its constituent MAPF queries. In particular, exRHCR employs a new extension of Priority-Based Search (PBS), a state-of-the-art MAPF solver.
Our extension, called exPBS, allows to warm-start the search with the priorities between agents used by PBS in the previous MAPF instances. We demonstrate empirically that exRHCR solves L-MAPF up to 25% faster than RHCR, and allows to increase throughput for given task streams by as much as 3%-16% by increasing the number of agents we can cope with for a given time budget.
You can see the seminar here
Work towards MSc degree under the supervision of Prof. Vadim Indelman
When: 24.1.2022 at 14:00
Where: zoom
Abstract: One of the most complex tasks of decision making and planning is to gather information. This task becomes even more complex when the state is high-dimensional and its belief cannot be expressed with a parametric distribution. Although the state is high-dimensional, in many problems only a small fraction of it might be involved in transitioning the state and generating observations. We exploit this fact to calculate an information-theoretic expected reward, mutual information (MI), over a much lower-dimensional subset of the state, to improve efficiency and without sacrificing accuracy. A similar approach was used in previous works, yet specifically for Gaussian distributions, and we here extend it for general distributions. Moreover, we apply the dimensionality reduction for cases in which the new states are augmented to the previous, yet again without sacrificing accuracy. We then continue by developing an estimator for the MI which works in a Sequential Monte Carlo (SMC) manner, and avoids the reconstruction of future belief’s surfaces. Finally, we show how this work is applied to the informative planning optimization problem. This work is then evaluated in a simulation of an active SLAM problem, where the improvement in both accuracy and timing is demonstrated.
You can see the seminar here
When: 26.1.2022 at 15:00
Where: zoom
Abstract: In this talk I will describe how formal methods such as synthesis – automatically creating a system from a formal specification – can be leveraged to design robots, and explain and provide guarantees for their behavior. I will discuss the benefits and challenges of synthesis techniques and will give examples of different robotic systems including modular robots, swarms and robots interacting with people.
You can see the seminar here
Work towards MSc degree under the supervision of Prof. Miriam Zacksenhouse
When: 25.1.2022 at 14:00
Where: zoom
Abstract: Over the years, industrial robots have been employed increasingly to perform advanced production and high-precision tasks in various industries. However, further integration of industrial robots is hampered by their lack of flexibility, adaptability, and decision-making capabilities compared to human operators. Contact-rich assembly tasks are incredibly challenging since even small uncertainties in the location of the parts can cause large reaction forces and prevent the robot from performing the task successfully. Large industries overcome this problem by designing precise assembly lines. However, this approach is very costly and not economical for small and medium industries where production volumes are not very large. An alternative approach is to use admittance control to handle location uncertainties by endowing the robot’s end-effector with proper stiffness, damping, and inertia properties to correct its position in response to the reaction forces. The power of admittance control for performing assembly tasks motivated a number of researchers to develop machine learning tools to tune the admittance parameters explicitly. However, learning was limited to diagonal admittance matrices and thus required learning the trajectory too. My thesis is based on the understanding that non-diagonal elements in the admittance matrices are critical for correcting the robot’s motion during assembly tasks. In particular, those elements can cause the robot to perform proper translation movements in response to reaction torques due to initial misalignments. In my Thesis, I used Reinforcement learning, and in particular, proximal policy optimization (PPO), to learn the parameters of non-diagonal admittance matrices for peg-in-hole assembly tasks. Learning was performed in simulations and tested in both simulations and on a real robot (UR5e). Results demonstrate that the learned admittance control is robust to uncertainties and can generalize to different locations and sizes. Interestingly, I show that the learned admittance matrices are space invariant. Most importantly, the learned policy was demonstrated successfully on a physical robot, for both rigid pegs and semi-rigid wires, without any retraining. This research is funded by Israel Innovation Authorities (IIA) as part of the ART (Assembly by Robotic Technology) academia-industry cooperation (“MAGNET”) aimed to develop generic tools for increasing robotic integration in the industry, especially for small to medium volumes.
You can see the seminar here
Work towards MSc degree under the supervision of Dr. Oren Salzman
When: 20.1.2022 at 11:00
Where: zoom
Abstract: We consider the problem of finding collision-free paths for curvature-constrained systems in the presence of obstacles while minimizing execution time. Specifically, we focus on the setting where a planar system can travel at some range of speeds with unbounded acceleration. This setting can model many systems, such as fixed-wing drones. Unfortunately, planning for such systems might require evaluating many (local) time-optimal transitions connecting two close-by configurations, which is computationally expensive. Existing methods either pre-compute all such transitions in a preprocessing stage or use heuristics to speed up the search, thus foregoing any guarantees on solution quality. Our key insight is that computing all the time-optimal transitions is both (i) computationally expensive and (ii) unnecessary for many problem instances. We show that by finding bounded-suboptimal solutions (solutions whose cost is bounded by 1+ε times the cost of the optimal solution for any user-provided ε) and not time-optimal solutions, one can dramatically reduce the number of time-optimal transitions used. We demonstrate using empirical evaluation that our planning framework can reduce the runtime by several orders of magnitude compared to the state-of-the-art while still providing guarantees on the quality of the solution.
You can see the seminar here