Ambarish Goswami: Safe fall strategies for humanoid robots

ambarish.com Ambarish Goswami's Current Research

Home/ Current Research/ Safe fall strategies for humanoid robots

Contact Info

Safe fall strategies for humanoid robots

Safety is a primary concern that must be addressed before humanoid robots can freely exist in human surroundings. Out of a number of possible situations where safety becomes an issue, one that involves a fall is particularly worrisome. Fall from an upright posture can cause damage to the robot, to delicate and expensive objects in the surrounding or to a human being. Regardless of the substantial progress in humanoid robot balance control strategies, the possibility of a fall remains real, even unavoidable. Yet, a comprehensive study of humanoid fall and prescribed fall strategies are rare.

One can ignore the possibility of a fall and wishfully hope that its effects will not be serious. However, failure studies, such as in car crash, have taught us against behaving according to this instinct. In fact, planning and simulation of failure situations can have enormous benefits, including system design improvements, and support for user safety and confidence. Following this philosophy we closely focus our attention to the phenomenon of humanoid fall and attempt to develop a comprehensive control strategy to deal with this undesired and traumatic ``failure'' event.

humanoid robot falls among multiple objects, no control

Humanoid robot falls among multiple objects. With no control, the robot nmay hit an object.

humanoid robot successfully avoids hitting object with fall controller
Humanoid robot can successfully avoid hitting surrounding objects using a fall controller.

A humanoid fall may be caused due to unexpected or excessive external forces, unusual or unknown slipperiness, slope or profile of the ground, causing the robot to slip, trip or topple. In these cases the disturbances that threaten balance are larger than what the balance controller can handle. Fall can also result from actuator, power or communication failure where the balance controller is partially or fully incapacitated. We currently consider only those situations in which the motor power is retained such that the robot can execute a prescribed control strategy.

A fall controller can target two major objectives independently or in combination: a) fall with a minimum damage and b) change fall direction such that the robot does not hit a certain object or person. Here we introduce a strategy for fall direction change and describe a controller which can achieve both objectives.

Let us note that a fall controller is not a balance controller. A fall controller complements, and does not replace, a balance controller. Only when the default balance controller has failed to stabilize the robot, the fall controller is activated. Further, a fall controller is not a push-recovery controller. A push-recovery controller is essentially a balance controller, which specifically deals with external disturbances of larger magnitude. A robot can recover from a push e.g., through an appropriate stepping strategy.

We propose a fall strategy which rapidly modifies the fall direction of a robot in order to avoid hitting a person or an object in the vicinity. Our approach is based on the optimal modification of the support base geometry of the robot through intelligent stepping. Additional improvement to the fall controller is achieved through inertia shaping technique aimed at controlling the centroidal rotational inertia of the robot. The video is composed of two simulation animations which show the falling motion of an Asimo-like humanoid robot. In the first simulation only the footstep controller is active. In the second animation, the inertia shaping controller is activated as soon as the humanoid makes a foot touchdown.

Humanoid Fall: Powerpoint presentation

Animations

Falling motion of NAO robot: No Control
Falling motion of NAO robot: Intelligent Stepping Control
Falling motion of NAO robot: Inertia Shaping Control

A list of my papers on this topic:

S-H. Lee and A. Goswami,
Fall on Backpack: Damage Minimizing Humanoid Fall on Targeted Body Segment Using Momentum Control,
ASME 2011 8th International Conference on Multibody Systems, Nonlinear Dynamics, and Control (MSNDC) inside International Design Engineering Technical Conference (IDETC), Washington DC, USA, August 2011.
(pdf).

Abstract:
Safety and robustness will become critical issues when humanoid robots start sharing human environments in the future. In physically interactive human environments, a catastrophic fall is the main threat to safety and smooth operation of humanoid robots, and thus it is critical to explore how to manage an unavoidable fall of humanoids. This paper deals with the problem of reducing the impact damage to a robot associated with a fall. A common approach is to employ damage-resistant design and apply impact-absorbing material to robot limbs, such as the backpack and knee, that are particularly prone to fall related impacts. In this paper, we select the backpack to be the most preferred body segment to experience an impact. We proceed to propose a control strategy that attempts to re-orient the robot during the fall such that it impacts the ground with its backpack. We show that the robot can fall on the backpack even when it starts falling sideways. This is achieved by utilizing dynamic coupling, i.e., by rotating the swing leg aiming to generate spin rotation of the trunk (backpack), and by rotating the trunk backward to drive the trunk to touch down with the backpack. The planning and control algorithms for fall are demonstrated in simulation.

S. Kalyanakrishnan and A. Goswami,
Learning to Predict Humanoid Fall,
The International Journal of Humanoid Robotics, Vol. 8, No. 2 (2011).
(pdf).

Abstract:
Falls are undesirable in humanoid robots, but also inevitable, especially as robots get deployed in physically interactive human environments. We consider the problem of fall prediction: to predict if the balance controller of a robot can prevent a fall from the robot’s current state. A trigger from the fall predictor is used to switch the robot from a balance maintenance mode to a fall control mode. It is desirable for the fall predictor to signal imminent falls with sufficient lead time before the actual fall, while minimizing false alarms. Analytical techniques and intuitive rules fail to satisfy these competing objectives on a large robot that is subjected to strong disturbances and exhibits complex dynamics. We contribute a novel approach to engineer fall data such that existing supervised learning methods can be exploited to achieve reliable prediction. Our method provides parameters to control the tradeoff between the false positive rate and lead time. Several combinations of parameters yield solutions that improve both the false positive rate and the lead time of hand-coded solutions. Learned solutions are decision lists with typical depths of 5–10, in a 16-dimensional feature space. Experiments are carried out in simulation on an ASIMO-like robot.

S. Kalyanakrishnan and A. Goswami,
Predicting falls of a humanoid robot through machine learning,
IAAI-10, Atlanta, Georgia, USA, July, 2010.
(pdf).

Abstract:
Although falls are undesirable in humanoid robots, they are also inevitable, especially as robots get deployed in physically interactive human environments. We consider the problem of "fall prediction", i.e., to predict if a robot's balance controller can prevent a fall from the current state. A trigger from the fall predictor is used to switch the robot from a balance maintenance mode to a fall control mode. Hence, it is desirable for the fall predictor to signal imminent falls with sufficient lead time before the actual fall, while minimizing false alarms. Analytical techniques and intuitive rules fail to satisfy these competing objectives on a large robot that is subjected to strong disturbances and exhibits complex dynamics.

Today effective supervised learning tools are available for finding patterns in high-dimensional data. Our paper contributes a novel approach to engineer fall data such that a supervised learning method can be exploited to achieve reliable prediction. Specifically, we introduce parameters to control the tradeoff between the false positive rate and lead time. Several parameter combinations yield solutions that improve both the false positive rate and the lead time of hand-coded solutions. Learned predictors are decision lists with typical depths of 5-10, in a 16-dimensional feature space. Experiments are carried out in simulation on an Asimo-like robot.

U. Nagarajan and A. Goswami,
Generalized Direction Changing Fall Control of Humanoid Robots Among Multiple Objects,
ICRA 2010, Anchorage, Alaska, May 2010.
(pdf).

Abstract:
Humanoid robots are expected to share human environments in the future and it is important to ensure safety of their operation. A serious threat to safety is the fall of a humanoid robot, which can seriously damage both the robot and objects in its surrounding. This paper proposes a strategy for planning and control of fall. The controller's objective is to prevent the robot from hitting surrounding objects during a fall by modifying its default fall direction.

We have earlier presented such a direction-changing fall controller, see here. However, the controller was applicable only when the robot's surrounding contained a single object. In this paper we introduce a generalized approach to humanoid fall-direction control among multiple objects. This new framework algorithmically establishes a desired fall direction through assigned scores, considers a number of control options, and selects and executes the best strategy. The fall planner is also able to select "No Action" as the best strategy, if appropriate. The controller is interactive and is applicable for fall occurring during upright standing or walking. The fall performance is continuously tracked and can be improved in real-time. The planning and control algorithms are demonstrated in simulation on an ASIMO-like humanoid robot.

Direction Changing Humanoid Fall

S.-K. Yun, A. Goswami and Y. Sakagami,
Safe Fall: Humanoid robot fall direction change through intelligent stepping and inertia shaping,
ICRA 2009, Kobe, Japan, May 2009.
(pdf).

Abstract:
Although fall is a rare event in the life of a humanoid robot, we must be prepared for it because its consequences are serious. In this paper we present a fall strategy which rapidly modifies the robot's fall direction in order to avoid hitting a person or an object in the vicinity. Our approach is based on the key observation that during "toppling" the rotational motion of a robot necessarily occurs at the leading edge or the leading corner of its support base polygon. To modify the fall direction the robot needs to change the position and orientation of this edge or corner vis-a-vis the prohibited direction. We achieve it through intelligent stepping as soon as a fall is detected. We compute the optimal stepping location which results in the safest fall. Additional improvement to the fall controller is achieved through inertia shaping techniques aimed at controlling the centroidal inertia of the robot.

We demonstrate our results through the simulation of an ASIMO-like humanoid robot. To our knowledge, this is the first implementation of a controller that attempts to change the fall direction of a humanoid robot.

Back to Current Research

Back to Main Page

Page last updated June 12, 2011