Planning, Learning and Reasoning Framework for Robot Truck Unloading

Islam, Fahad; Vemula, Anirudh; Kim, Sung-Kyun; Dornbush, Andrew; Salzman, Oren; Likhachev, Maxim

doi:10.1109/icra40945.2020.9196604

“…Once the high-level decision is made, we use other heuristic to determine the exact pick location or the sweep distance. We refer to our previous work [1] for details on how a high-level decision is executed for this Truck Unloading problem. Input Features: We simulate perception sensors such that the high-level decision is based only based on simulated perception data and not ground-truth information from the simulator.…”

Section: Experiments and Resultsmentioning

confidence: 99%

“…This introduces partial observability of the complete state and singularities where the inverse map from compressed state to a full state state is not unique. As shown in our previous work [1], to handle uncertainty, a standard approach would be to compute a probabilistic distribution of the compressed state (beliefstate), and formulate the planning problem as solving a Belief Markov Decision Process [6]. However, we observed that the amount of data required for credit assignment and sequential reasoning is relatively very large.…”

Section: Related Workmentioning

confidence: 87%

See 1 more Smart Citation

Learning Optimal Decision Making for an Industrial Truck Unloading Robot using Minimal Simulator Runs

Das,

Vemula,

Pathak

et al. 2021

Preprint

Self Cite

0

View full text Add to dashboard Cite

Consider a truck filled with boxes of varying size and unknown mass and an industrial robot with end-effectors that can unload multiple boxes from any reachable location. In this work, we investigate how would the robot with the help of a simulator, learn to maximize the number of boxes unloaded by each action. Most high-fidelity robotic simulators like ours are time-consuming. Therefore, we investigate the above learning problem with a focus on minimizing the number of simulation runs required. The optimal decision-making problem under this setting can be formulated as a multi-class classification problem. However, to obtain the outcome of any action requires us to run the time-consuming simulator, thereby restricting the amount of training data that can be collected. Thus, we need a data-efficient approach to learn the classifier and generalize it with a minimal amount of data. A high-fidelity physics-based simulator is common in general for complex manipulation tasks involving multi-body interactions. To this end, we train an optimal decision tree as the classifier, and for each branch of the decision tree, we reason about the confidence in the decision using a Probably Approximately Correct (PAC) framework to determine whether more simulator data will help reach a certain confidence level. This provides us with a mechanism to evaluate when simulation can be avoided for certain decisions, and when simulation will improve the decision making. For the truck unloading problem, our experiments show that a significant reduction in simulator runs can be achieved using the proposed method as compared to naively running the simulator to collect data to train equally performing decision trees. I. INTRODUCTIONMany robotics applications require planning and decision making based on what the robot observes in order to complete a task. In this article we study the problem of robotic truck unloading, where the task is to empty the truck by unloading all the boxes. Fig. 1 shows an industrial truck unloading robot and cardboard boxes inside the truck that needs to be unloaded. The robot has two end effectors, one is fixed to the base of the robot and can sweep boxes from the floor, and another suspended with an arm and can pick boxes using a plunger mechanism. The task of emptying the truck can be broken down into small sub-tasks like picking boxes from a certain location or sweeping boxes from the floor. This problem involves both task planning and motion planning. We exploit the fact that our problem allows us to perform task planning and motion planning independently, and hence in this paper, we focus only on the task planning.

show abstract

“…Motion primitives, introduced by Frazzoli et al [17], have been used in many motion planners [23,24,33,42,35]. In our setting, the motion primitives are a set of predefined kinematically feasible local motions.…”

Section: B Motion Primitivesmentioning

confidence: 99%

Toward Certifiable Motion Planning for Medical Steerable Needles

Fu¹,

Salzman²,

Alterovitz³

2021

Robotics: Science and Systems XVII

Self Cite

View full text Add to dashboard Cite

Medical steerable needles can move along 3D curvilinear trajectories to avoid anatomical obstacles and reach clinically significant targets inside the human body. Automating steerable needle procedures can enable physicians and patients to harness the full potential of steerable needles by maximally leveraging their steerability to safely and accurately reach targets for medical procedures such as biopsies and localized therapy delivery for cancer. For the automation of medical procedures to be clinically accepted, it is critical from a patient care, safety, and regulatory perspective to certify the correctness and effectiveness of the motion planning algorithms involved in procedure automation. In this paper, we take an important step toward creating a certifiable motion planner for steerable needles. We introduce the first motion planner for steerable needles that offers a guarantee, under clinically appropriate assumptions, that it will, in finite time, compute an exact, obstacle-avoiding motion plan to a specified target, or notify the user that no such plan exists. We present an efficient, resolution-complete motion planner for steerable needles based on a novel adaptation of multi-resolution planning. Compared to state-of-the-art steerable needle motion planners (none of which provide any completeness guarantees), we demonstrate that our new resolution-complete motion planner computes plans faster and with a higher success rate.

show abstract

“…Motion primitives, introduced by Frazzoli et al [17], have been used in many motion planners [23,24,33,42,35]. In our setting, the motion primitives are a set of predefined kinematically feasible local motions.…”

Section: B Motion Primitivesmentioning

confidence: 99%

Toward Certifiable Motion Planning for Medical Steerable Needles

Fu,

Salzman,

Alterovitz

2021

Preprint

Self Cite

View full text Add to dashboard Cite

Medical steerable needles can move along 3D curvilinear trajectories to avoid anatomical obstacles and reach clinically significant targets inside the human body. Automating steerable needle procedures can enable physicians and patients to harness the full potential of steerable needles by maximally leveraging their steerability to safely and accurately reach targets for medical procedures such as biopsies and localized therapy delivery for cancer. For the automation of medical procedures to be clinically accepted, it is critical from a patient care, safety, and regulatory perspective to certify the correctness and effectiveness of the motion planning algorithms involved in procedure automation. In this paper, we take an important step toward creating a certifiable motion planner for steerable needles. We introduce the first motion planner for steerable needles that offers a guarantee, under clinically appropriate assumptions, that it will, in finite time, compute an exact, obstacle-avoiding motion plan to a specified target, or notify the user that no such plan exists. We present an efficient, resolution-complete motion planner for steerable needles based on a novel adaptation of multi-resolution planning. Compared to state-of-the-art steerable needle motion planners (none of which provide any completeness guarantees), we demonstrate that our new resolution-complete motion planner computes plans faster and with a higher success rate.

show abstract

Planning, Learning and Reasoning Framework for Robot Truck Unloading

Cited by 9 publications

References 16 publications

Learning Optimal Decision Making for an Industrial Truck Unloading Robot using Minimal Simulator Runs

Learning Optimal Decision Making for an Industrial Truck Unloading Robot using Minimal Simulator Runs

Toward Certifiable Motion Planning for Medical Steerable Needles

Toward Certifiable Motion Planning for Medical Steerable Needles

Contact Info

Product

Resources

About