# Underactuated Robotics

Algorithms for Walking, Running, Swimming, Flying, and Manipulation

Russ Tedrake

How to cite these notes   |   Send me your feedback

Note: These are working notes used for a course being taught at MIT. They will be updated throughout the Spring 2020 semester. Lecture videos are available on YouTube.

# Deriving the equations of motion (an example)

The equations of motion for a standard robot can be derived using the method of Lagrange. Using $T$ as the total kinetic energy of the system, and $U$ as the total potential energy of the system, $L = T-U$, and $Q_i$ as the generalized force corresponding to $q_i$, the Lagrangian dynamic equations are: $$\frac{d}{dt}\pd{L}{\dot{q}_i} - \pd{L}{q_i} = Q_i.$$ If you are not comfortable with these equations, then any good book chapter on rigid body mechanics can bring you up to speed -- try Craig89 for a very practical guide to robot kinematics/dynamics, Goldstein02 for a more hard-core dynamics text or Thornton03 for a classical dynamics text which is a nice read -- for now you can take them as a handle that you can crank to generate equations of motion.

# Simple Double Pendulum

Consider the simple double pendulum with torque actuation at both joints and all of the mass concentrated in two points (for simplicity). Using $\bq = [\theta_1,\theta_2]^T$, and ${\bf p}_1,{\bf p}_2$ to denote the locations of $m_1,m_2$, respectively, the kinematics of this system are: \begin{eqnarray*} {\bf p}_1 =& l_1\begin{bmatrix} s_1 \\ - c_1 \end{bmatrix}, &{\bf p}_2 = {\bf p}_1 + l_2\begin{bmatrix} s_{1+2} \\ - c_{1+2} \end{bmatrix} \\ \dot{{\bf p}}_1 =& l_1 \dot{q}_1\begin{bmatrix} c_1 \\ s_1 \end{bmatrix}, &\dot{{\bf p}}_2 = \dot{{\bf p}}_1 + l_2 (\dot{q}_1+\dot{q}_2) \begin{bmatrix} c_{1+2} \\ s_{1+2} \end{bmatrix} \end{eqnarray*} Note that $s_1$ is shorthand for $\sin(q_1)$, $c_{1+2}$ is shorthand for $\cos(q_1+q_2)$, etc. From this we can write the kinetic and potential energy: \begin{align*} T =& \frac{1}{2} m_1 \dot{\bf p}_1^T \dot{\bf p}_1 + \frac{1}{2} m_2 \dot{\bf p}_2^T \dot{\bf p}_2 \\ =& \frac{1}{2}(m_1 + m_2) l_1^2 \dot{q}_1^2 + \frac{1}{2} m_2 l_2^2 (\dot{q}_1 + \dot{q}_2)^2 + m_2 l_1 l_2 \dot{q}_1 (\dot{q}_1 + \dot{q}_2) c_2 \\ U =& m_1 g y_1 + m_2 g y_2 = -(m_1+m_2) g l_1 c_1 - m_2 g l_2 c_{1+2} \end{align*} Taking the partial derivatives $\pd{T}{q_i}$, $\pd{T}{\dot{q}_i}$, and $\pd{U}{q_i}$ ($\pd{U}{\dot{q}_i}$ terms are always zero), then $\frac{d}{dt}\pd{T}{\dot{q}_i}$, and plugging them into the Lagrangian, reveals the equations of motion: \begin{align*} (m_1 + m_2) l_1^2 \ddot{q}_1& + m_2 l_2^2 (\ddot{q}_1 + \ddot{q}_2) + m_2 l_1 l_2 (2 \ddot{q}_1 + \ddot{q}_2) c_2 \\ &- m_2 l_1 l_2 (2 \dot{q}_1 + \dot{q}_2) \dot{q}_2 s_2 + (m_1 + m_2) l_1 g s_1 + m_2 g l_2 s_{1+2} = \tau_1 \\ m_2 l_2^2 (\ddot{q}_1 + \ddot{q}_2)& + m_2 l_1 l_2 \ddot{q}_1 c_2 + m_2 l_1 l_2 \dot{q}_1^2 s_2 + m_2 g l_2 s_{1+2} = \tau_2 \end{align*} As we saw in chapter 1, numerically integrating (and animating) these equations in produces the expected result.

# The Manipulator Equations

If you crank through the Lagrangian dynamics for a few simple robotic manipulators, you will begin to see a pattern emerge - the resulting equations of motion all have a characteristic form. For example, the kinetic energy of your robot can always be written in the form: $$T = \frac{1}{2} \dot{\bq}^T \bM(\bq) \dot{\bq},$$ where $\bM$ is the state-dependent inertia matrix (aka mass matrix). This observation affords some insight into general manipulator dynamics - for example we know that ${\bf M}$ is always positive definite, and symmetricAsada86(p.107) and has a beautiful sparsity patternFeatherstone05 that we'll attempt to take advantage of in our algorithms.

Continuing our abstractions, we find that the equations of motion of a general robotic manipulator (without kinematic loops) take the form $${\bf M}(\bq)\ddot{\bq} + \bC(\bq,\dot{\bq})\dot{\bq} = \btau_g(\bq) + {\bf B}\bu, \label{eq:manip}$$ where $\bq$ is the joint position vector, ${\bf M}$ is the inertia matrix, $\bC$ captures Coriolis forces, and $\btau_g$ is the gravity vector. The matrix $\bB$ maps control inputs $\bu$ into generalized forces. Note that we pair $\bM\ddot\bq + \bC\dot\bq$ on the left side because "... the equations of motion depend on the choice of coordinates $\bq$. For this reason neither $\bM\ddot\bq$ nor $\bC\dot\bq$ individually should be thought of as a generalized force; only their sum is a force"Choset05(s.10.2). Indeed, whenever I write Eq. (\ref{eq:manip}), I see $ma = F$.

# Manipulator Equation form of the Simple Double Pendulum

The equations of motion from Example 1 can be written compactly as: \begin{align*} \bM(\bq) =& \begin{bmatrix} (m_1 + m_2)l_1^2 + m_2 l_2^2 + 2 m_2 l_1l_2 c_2 & m_2 l_2^2 + m_2 l_1 l_2 c_2 \\ m_2 l_2^2 + m_2 l_1 l_2 c_2 & m_2 l_2^2 \end{bmatrix} \\ \bC(\bq,\dot\bq) =& \begin{bmatrix} 0 & -m_2 l_1 l_2 (2\dot{q}_1 + \dot{q}_2)s_2 \\ m_2 l_1 l_2 \dot{q}_1 s_2 & 0 \end{bmatrix} \\ \btau_g(\bq) =& -g \begin{bmatrix} (m_1 + m_2) l_1 s_1 + m_2 l_2 s_{1+2} \\ m_2 l_2 s_{1+2} \end{bmatrix} , \quad \bB = \begin{bmatrix} 1 & 0 \\ 0 & 1 \end{bmatrix} \end{align*} Note that this choice of the $\bC$ matrix was not unique.

The manipulator equations are very general, but they do define some important characteristics. For example, $\ddot{\bq}$ is (state-dependent) linearly related to the control input, $\bu$. This observation justifies the control-affine form of the dynamics assumed throughout the notes.

Note that we have chosen to use the notation of second-order systems (with $\dot{\bq}$ and $\ddot{\bq}$ appearing in the equations) throughout this book. Although I believe it provides more clarity, there is an important limitation to this notation: it is impossible to describe 3D rotations in "minimal coordinates" using this notation without introducing kinematic singularities (like the famous "gimbal lock"). For instance, a common singularity-free choice for representing a 3D rotation is the unit quaternion, described by 4 real values (plus a norm constraint). However we can still represent the rotational velocity without singularities using just 3 real values. This means that the length of our velocity vector is no longer the same as the length of our position vector. For this reason, you will see that most of the software in uses the more general notation with $\bv$ to represent velocity, $\bq$ to represent positions, and the manipulator equations are written as $$\bM(\bq) \dot{\bv} + \bC(\bq,\bv)\bv = \btau_g(\bq) + \bB \bu,$$ which is not necessarily a second-order system. See Duindam06 for a nice discussion of this topic.

# Recursive Dynamics Algorithms

The equations of motions for our machines get complicated quickly. Fortunately, for robots with a tree-link kinematic structure, there are very efficient and natural recursive algorithms for generating these equations of motion. For a detailed reference on these methods, see Featherstone07; some people prefer reading about the Articulated Body Method in Mirtich96. The implementation in uses a related formulation from Jain11.

# Bilateral Position Constraints

If our robot has closed-kinematic chains, for instance those that arise from a four-bar linkage, then we need a little more. The Lagrangian machinery above assumes "minimal coordinates"; if our state vector $\bq$ contains all of the links in the kinematic chain, then we do not have a minimal parameterization -- each kinematic loop adds (at least) one constraint so should remove (at least) one degree of freedom. Although some constraints can be solved away, the more general solution is to use the Lagrangian to derive the dynamics of the unconstrained system (a kinematic tree without the closed-loop constraints), then add additional generalized forces that ensure that the constraint is always satisfied.

Consider the constraint equation $$\bh(\bq) = 0.$$ For the case of the kinematic closed-chain, this can be the kinematic constraint that the location of one end of the chain is equal to the location of the other end of the chain. The equations of motion can be written $$\bM({\bq})\ddot{\bq} + \bC(\bq,\dot{\bq})\dot\bq = \btau_g(\bq) + \bB\bu + \bH^T(\bq) \blambda,\label{eq:constrained_manip}$$ where $\bH(\bq) = \pd\bh{\bq}$ and ${\blambda}$ is the constraint force. Let's use the shorthand $$\bM({\bq})\ddot{\bq} = \btau(q,\dot{q},u) + \bH^T(\bq) \blambda. \label{eq:manip_short}$$

Using \begin{gather}\dot\bh = \bH\dot\bq,\\ \ddot\bh = \bH \ddot\bq + \dot\bH \dot\bq, \label{eq:ddoth} \end{gather} we can solve for $\blambda$, by observing that when the constraint is imposed, $\bh=0$ and therefore $\dot\bh=0$ and $\ddot\bh=0$. Inserting the dynamics (\ref{eq:manip_short}) into (\ref{eq:ddoth}) yields $$\blambda =- (\bH \bM^{-1} \bH^T)^+ (\bH \bM^{-1} \btau + \dot\bH\dot\bq).\label{eq:lambda_from_h}$$ The $^+$ notation refers to a Moore-Penrose pseudo-inverse. In many cases, this matrix will be full rank (for instance, in the case of multiple independent four-bar linkages) and the traditional inverse could have been used. When the matrix drops rank (multiple solutions), then the pseudo-inverse will select the solution with the smallest constraint forces in the least-squares sense.

To combat numerical "constraint drift", one might like to add a restoring force in the event that the constraint is not satisfied to numerical precision. To accomplish this, rather than solving for $\ddot\bh = 0$ as above, we can instead solve for $$\ddot\bh = \bH\ddot\bq + \dot\bH\dot\bq = -2\alpha \dot\bh - \alpha^2 \bh,$$ where $\alpha>0$ is a stiffness parameter. This is known as Baumgarte's stabilization technique, implemented here with a single parameter to provide a critically-damped response. Carrying this through yields $$\blambda =- (\bH \bM^{-1} \bH^T)^+ (\bH \bM^{-1} \btau + (\dot{\bH} + 2\alpha \bH)\dot{\bq} + \alpha^2 \bh).$$

There is an important optimization-based derivation/interpretation of these equations, which we will build on below, using Gauss's Principle of Least Constraint. This principle says that we can derive the constrained dynamics as : \begin{align} \min_\ddot{\bq} \quad & \frac{1}{2} (\ddot{\bq} - \ddot{\bq}_{uc})^T \bM (\ddot{\bq} - \ddot{\bq}_{uc}), \label{eq:least_constraint} \\ \subjto \quad & \ddot{\bh}(\bq,\dot\bq,\ddot\bq) = 0 = \bH \ddot{\bq} + \dot\bH \dot{\bq}, \nonumber \end{align} where $\ddot{\bq}_{uc} = \bM^{-1}\tau$ is the "unconstrained acceleration" Udwadia92. Equation \eqref{eq:constrained_manip} comes right out of the optimality conditions with $\lambda$ acting precisely as the Lagrange multiplier. This is a convex quadratic program with equality constraints, which has a closed-form solution that is precisely Equation \eqref{eq:lambda_from_h} above. It is also illustrative to look at the dual formulation of this optimization, which can be written as $$\min_\lambda \frac{1}{2} \lambda^T \bH \bM^{-1} \bH^T \lambda - \lambda^T (\bH \bM^{-1}\btau + \dot\bH \dot{\bq}).$$ Observe that the linear term is contains the acceleration of $\bh$ if the dynamics evolved without the constraint forces applied; let's call that $\ddot\bh_{uc}$: $$\min_\lambda \frac{1}{2} \lambda^T \bH \bM^{-1} \bH^T \lambda - \lambda^T \ddot{\bh}_{uc}.$$ The primal formulation has accelerations as the decision variables, and the dual formulation has constraint forces as the decision variables. For now this is merely a cute observation; we'll build on it more when we get to discussing contact forces.

# Bilateral Velocity Constraints

Consider the constraint equation $$\bh_v(\bq,\dot{\bq}) = 0,$$ where $\pd{\bh_v}{\dot{\bq}} \ne 0.$ These are less common, but arise when, for instance, a joint is driven through a prescribed motion. Here, the manipulator equations are given by $$\bM\ddot{\bq} = \btau + \pd{\bh_v}{\dot{\bq}}^T \blambda.$$ Using $$\dot\bh_v = \pd{\bh_v}{\bq} \dot{\bq} + \pd{\bh_v}{\dot{\bq}}\ddot{\bq},$$ setting $\dot\bh_v = 0$ yields $$\blambda = - \left( \pd{\bh_v}{\dot{\bq}} \bM^{-1} \pd{\bh_v}{\dot{\bq}} \right)^+ \left[ \pd{\bh_v}{\dot{\bq}} \bM^{-1} \btau + \pd{\bh_v}{\bq} \dot{\bq} \right].$$

Again, to combat constraint drift, we can ask instead for $\dot\bh_v = -\alpha \bh_v$, which yields $$\blambda = - \left( \pd{\bh_v}{\dot{\bq}} \bM^{-1} \pd{\bh_v}{\dot{\bq}} \right)^+ \left[ \pd{\bh_v}{\dot{\bq}} \bM^{-1} \btau + \pd{\bh_v}{\bq} \dot{\bq} + \alpha \bh_v \right].$$

# The Dynamics of Contact

The dynamics of multibody systems that make and break contact are closely related to the dynamics of constrained systems, but tend to be much more complex. In the simplest form, you can think of non-penetration as an inequality constraint: the signed distance between collision bodies must be non-negative. But, as we have seen in the chapters on walking, the transitions when these constraints become active correspond to collisions, and for systems with momentum they require some care. We'll also see that frictional contact adds its own challenges.

There are three main approaches used to modeling contact with "rigid" body systems: 1) rigid contact approximated with stiff compliant contact, 2) hybrid models with collision event detection, impulsive reset maps, and continuous (constrained) dynamics between collision events, and 3) rigid contact approximated with time-averaged forces (impulses) in a time-stepping scheme. Each modeling approach has advantages/disadvantages for different applications.

Before we begin, there is a bit of notation that we will use throughout. Let $\phi(\bq)$ indicate the relative (signed) distance between two rigid bodies. For rigid contact, we would like to enforce the unilateral constraint: $$\phi(\bq) \geq 0.$$ We'll use ${\bf n} = \pd{\phi}{\bq}$ to denote the "contact normal", as well as ${\bf t}_1$ and ${\bf t}_2$ as a basis for the tangents at the contact (orthonormal vectors in the Cartesian space, projected into joint space), all represented as row vectors in joint coordinates.

We will also find it useful to assemble the contact normal and tangent vectors into a single matrix, $\bJ$, that we'll call the contact Jacobian: $$\bJ(\bq) = \begin{bmatrix} {\bf n}\\ {\bf t}_1\\ {\bf t}_2 \end{bmatrix}.$$ As written, $\bJ\bv$ gives the relative velocity of the closest points in the contact coordinates; it can be also be extended with three more rows to output a full spatial velocity (e.g. when modeling torsional friction). The generalized force produced by the contact is given by $\bJ^T \lambda$, where $\lambda = [f_n, f_{t1}, f_{t2}]^T:$ $$\bM(\bq) \dot{\bv} + \bC(\bq,\bv)\bv = \btau_g(\bq) + \bB \bu + \bJ^T(\bq)\lambda.$$

# Compliant Contact Models

Most compliant contact models are conceptually straight-forward: we will implement contact forces using a stiff spring (and damperHunt75) that produces forces to resist penetration (and grossly model the dissipation of collision and/or friction). For instance, for the normal force, $f_n$, we can use a simple (piecewise) linear spring law:

Coulomb friction is described by two parameters -- $\mu_{static}$ and $\mu_{dynamic}$ -- which are the coefficients of static and dynamic friction. When the contact tangential velocity (given by ${\bf t}\bv$) is zero, then friction will produce whatever force is necessary to resist motion, up to the threshold $\mu_{static} f_n.$ But once this threshold is exceeding, we have slip (non-zero contact tangential velocity); during slip friction will produce a constant drag force $|f_t| = \mu_{dynamic} f_n$ in the direction opposing the motion. This behaviour is not a simple function of the current state, but we can approximate it with a continuous function as pictured below.

With these two laws, we can recover the contact forces as relatively simple functions of the current state. However, the devil is in the details. There are a few features of this approach that can make it numerically unstable. If you've ever been working in a robotics simulator and watched your robot take a step only to explode out of the ground and go flying through the air, you know what I mean.

In order to tightly approximate the (nearly) rigid contact that most robots make with the world, the stiffness of the contact "springs" must be quite high. For instance, I might want my 180kg humanoid robot model to penetrate into the ground no more than 1mm during steady-state standing. The challenge with adding stiff springs to our model is that this results in stiff differential equations (it is not a coincidence that the word stiff is the common term for this in both springs and ODEs). As a result, the best implementations of compliant contact for simulation make use of special-purpose numerical integration recipes (e.g. Castro20) and compliant contact models are often difficult to use in e.g. trajectory/feedback optimization.

But there is another serious numerical challenge with the basic implementation of this model. Computing the signed-distance function, $\phi(\bq)$, when it is non-negative is straightforward, but robustly computing the "penetration depth" once two bodies are in collision is not. Consider the case of use $\bphi(\bq)$ to represent the maximum penetration depth of, say, two boxes that are contacting each other. With compliant contact, we must have some penetration to produce any contact force. But the direction of the normal vector, ${\bf n} = \pd{\bphi}{\bq},$ can easily change discontinuously as the point of maximal penetration moves, as I've tried to illustrate here:

If you really think about this example, it actually even more of the foibles of trying to even define the contact points and normals consistently. It seems reasonable to define the point on body B as the point of maximal penetration into body A, but notice that as I've drawn it, that isn't actually unique! The corresponding point on body A should probably be the point on the surface with the minimal distance from the max penetration point (other tempting choices would likely cause the distance to be discontinuous at the onset of penetration). But this whole strategy is asymmetric; why shouldn't I have picked the vector going with maximal penetration into body B? My point is simply that these cases exist, and that even our best software implementations get pretty unsatisfying when you look into the details. And in practice, it's a lot to expect the collision engine to give consistent and smooth contact points out in every situation.

Some of the advantages of this approach include (1) the fact that it is easy to implement, at least for simple geometries, (2) by virtue of being a continuous-time model it can be simulated with error-controlled integrators, and (3) the tightness of the approximation of "rigid" contact can be controlled through relatively intuitive parameters. However, the numerical challenges of dealing with penetration are very real, and they motivate our other two approaches that attempt to more strictly enforce the non-penetration constraints.

# Impulsive Collisions

The collision event is described by the zero-crossings (from positive to negative) of $\phi$. Let us start by assuming frictionless collisions, allowing us to write $$\bM\ddot{\bq} = \btau + \lambda {\bf n}^T,$$ where ${\bf n}$ is the "contact normal" we defined above and $\lambda \ge 0$ is now a (scalar) impulsive force that is well-defined when integrated over the time of the collision (denoted $t_c^-$ to $t_c^+$). Integrate both sides of the equation over that (instantaneous) interval: \begin{align*}\int_{t_c^-}^{t_c^+} dt\left[\bM \ddot{\bq} \right] = \int_{t_c^-}^{t_c^+} dt \left[ \btau + \lambda {\bf n}^T \right] \end{align*} Since $\bM$, $\btau$, and ${\bf n}$ are constants over this interval, we are left with $$\bM\dot{\bq}^+ - \bM\dot{\bq}^- = {\bf n}^T \int_{t_c^-}^{t_c^+} \lambda dt,$$ where $\dot{\bq}^+$ is short-hand for $\dot{\bq}(t_c^+)$. Multiplying both sides by ${\bf n} \bM^{-1}$, we have $${\bf n} \dot{\bq}^+ - {\bf n}\dot{\bq}^- = {\bf n} \bM^{-1} {\bf n}^T \int_{t_c^-}^{t_c^+} \lambda dt.$$ After the collision, we have $\dot\phi^+ = -e \dot\phi^-$, with $0 \le e \le 1$ denoting the coefficient of restitution, yielding: $${\bf n} \dot{\bq}^+ - {\bf n}\dot{\bq}^- = -(1+e){\bf n}\dot{\bq}^-,$$ and therefore $$\int_{t_c^-}^{t_c^+} \lambda dt = - (1+e)\left[ {\bf n} \bM^{-1} {\bf n}^T \right]^\# {\bf n} \dot{\bq}^-.$$ I've used the notation $A^\#$ here to denote the pseudoinverse of $A$ (I would normally write $A^+,$ but have changed it for this section to avoid confusion). Substituting this back in above results in $$\dot{\bq}^+ = \left[ \bI - (1+e)\bM^{-1} {\bf n}^T \left[{\bf n} \bM^{-1} {\bf n}^T \right]^\# {\bf n} \right] \dot{\bq}^-.$$

We can add friction at the contact. Rigid bodies will almost always experience contact at only a point, so we typically ignore torsional friction Featherstone07, and model only tangential friction by extending ${\bf n}$ to a matrix $$\bJ = \begin{bmatrix} {\bf n}\\ {\bf t}_1\\ {\bf t}_2 \end{bmatrix},$$ to capture the gradient of the contact location tangent to the contact surface. Then $\blambda$ becomes a Cartesian vector representing the contact impulse, and (for infinite friction) the post-impact velocity condition becomes ${\bf J}\dot{\bq}^+ = \text{diag}(-e, 0, 0) {\bf J}\dot{\bq}^-,$ resulting in the equations: $$\dot{\bq}^+ = \left[ \bI - \bM^{-1} \bJ^T \text{diag}(1+e, 1, 1) \left[\bJ \bM^{-1} \bJ^T \right]^\# \bJ \right]\dot{\bq}^-.$$ If $\blambda$ is restricted to a friction cone, as in Coulomb friction, in general we have to solve an optimization to solve for $\dot\bq^+$ subject to the inequality constraints.

# A spinning ball bouncing on the ground.

Imagine a ball (a hollow-sphere) in the plane with mass $m$, radius $r$. The configuration is given by $q = \begin{bmatrix} x, z, \theta \end{bmatrix}^T.$ The equations of motion are $$\bM(\bq)\ddot\bq = \begin{bmatrix} m & 0 & 0 \\ 0 & m & 0 \\ 0 & 0 & \frac{2}{3}mr^2 \end{bmatrix} \ddot\bq = \begin{bmatrix} 0 \\ -g \\ 0 \end{bmatrix} + \begin{bmatrix} 0 & 1 \\ 1 & 0 \\ 0 & r \end{bmatrix} \begin{bmatrix} \lambda_z \\ \lambda_x \end{bmatrix} = \tau_g + \bJ^T {\bf \lambda},$$ where ${\bf \lambda}$ are the contact forces (which are zero except during the impulsive collision). Given a coefficient of restitution $e$ and a collision with a horizontal ground, the post-impact velocities are: $$\dot{\bq}^+ = \begin{bmatrix} \frac{3}{5} & 0 & -\frac{2}{5} r \\ 0 & - e & 0 \\ -\frac{3}{5r} & 0 & \frac{2}{5}\end{bmatrix}\dot{\bq}^-.$$

# Putting it all together

We can put the bilateral constraint equations and the impulsive collision equations together to implement a hybrid model for unilateral constraints of the form. Let us generalize $$\bphi(\bq) \ge 0,$$ to be the vector of all pairwise (signed) distance functions between rigid bodies; this vector describes all possible contacts. At every moment, some subset of the contacts are active, and can be treated as a bilateral constraint ($\bphi_i=0$). The guards of the hybrid model are when an inactive constraint becomes active ($\bphi_i>0 \rightarrow \bphi_i=0$), or when an active constraint becomes inactive ($\blambda_i>0 \rightarrow \blambda_i=0$ and $\dot\phi_i > 0$). Note that collision events will (almost always) only result in new active constraints when $e=0$, e.g. we have pure inelastic collisions, because elastic collisions will rarely permit sustained contact.

If the contact involves Coulomb friction, then the transitions between sticking and sliding can be modeled as additional hybrid guards.

# Time-stepping Approximations for Rigid Contact

So far we have seen two different approaches to enforcing the inequality constraints of contact, $\phi(\bq) \ge 0$ and the friction cone. First we introduced compliant contact which effectively enforces non-penetration with a stiff spring. Then we discussed event detection as a means to switch between different models which treat the active constraints as equality constraints. But, perhaps surprisingly, one of the most popular and scalable approaches is something different: it involves formulating a mathematical program that can solve the inequality constraints directly on each time step of the simulation. These algorithms may be more expensive to compute on each time step, but they allow for stable simulations with potentially much larger and more consistent time steps.

# Complementarity formulations

What class of mathematical program due we need to simulate contact? The discrete nature of contact suggests that we might need some form of combinatorial optimization. Indeed, the most common transcription is to a Linear Complementarity Problem (LCP) Murty88, as introduced by Stewart96 and Anitescu97. An LCP is typically written as \begin{align*} \find_{\bw,\bz} \quad \subjto \quad & \bw = \bq + \bM \bz, \\ & \bw \ge 0, \bz \ge 0, \bw^T\bz = 0.\end{align*} They are directly related to the optimality conditions of a (potentially non-convex) quadratic program and Anitescu97> showed that the LCP's generated by our rigid-body contact problems can be solved by Lemke's algorithm. As short-hand we will write these complementarity constraints as $$\find_{\bz}\quad \subjto \quad 0 \le (\bq + \bM\bz) \perp \bz \ge 0.$$

Rather than dive into the full transcription, which has many terms and can be relatively difficult to parse, let me start with two very simple examples.

# Time-stepping LCP: Normal force

Consider our favorite simple mass being actuated by a horizontal force (with the double integrator dynamics), but this time we will add a wall that will prevent the cart position from taking negative values: our non-penetration constraint is $q \ge 0$. Physically, this constraint is implemented by a normal (horizontal) force, $f$, yielding the equations of motion: $$m\ddot{q} = u + f.$$

$f$ is defined as the force required to enforce the non-penetration constraint; certainly the following are true: $f \ge 0$ and $q \cdot f = 0$. $q \cdot f = 0$ is the "complementarity constraint", and you can read it here as "either q is zero or force is zero" (or both); it is our "no force at a distance" constraint, and it is clearly non-convex. It turns out that satisfying these constraints, plus $q \ge 0$, is sufficient to fully define $f$.

To define the LCP, we first discretize time, by approximating \begin{gather*}q[n+1] = q[n] + h v[n+1], \\ v[n+1] = v[n] + \frac{h}{m}(u[n] + f[n]).\end{gather*} This is almost the standard Euler approximation, but note the use of $v[n+1]$ in the right-hand side of the first equation -- this is actually a semi-implicit Euler approximation, and this choice is essential in the derivation.

Given $h, q[n], v[n],$ and $u[n]$, we can solve for $f[n]$ and $q[n+1]$ simultaneously, by solving the following LCP: \begin{gather*} q[n+1] = \left[ q[n] + h v[n] + \frac{h^2}{m} u[n] \right] + \frac{h^2}{m} f[n] \\ q[n+1] \ge 0, \quad f[n] \ge 0, \quad q[n+1]\cdot f[n] = 0. \end{gather*} Take a moment and convince yourself that this fits into the LCP prescription given above.

Note: Please don't confuse this visualization with the more common visualization of the solution space for an LCP (in two or more dimensions) in terms of "complementary cones"Murty88.

Perhaps it's also helpful to plot the solution, $q[n+1], f[n]$ as a function of $q[n], v[n]$. I've done it here for $m=1, h=0.1, u[n]=0$:

In the (time-stepping, "impulse-velocity") LCP formulation, we write the contact problem in it's combinatorial form. In the simple example above, the complementarity constraints force any solution to lie on either the positive x-axis ($f[n] \ge 0$) or the positive y-axis ($q[n+1] \ge 0$). The equality constraint is simply a line that will intersect with this constraint manifold at the solution. But in this frictionless case, it is important to realize that these are simply the optimality conditions of a convex optimization problem: the discrete-time version of Gauss's principle that we used above. Using $\bv'$ as shorthand for $\bv[n+1]$, and replacing $\ddot{\bq} = \frac{\bv' - \bv}{h}$ in Eq \eqref{eq:least_constraint} and scaling the objective by $h^2$ we have: \begin{align*} \min_{\bv'} \quad & \frac{1}{2} \left(\bv' - \bv - h\bM^{-1}\btau\right)^T \bM \left(\bv' - \bv - h\bM^{-1}\btau\right) \\ \subjto \quad & \frac{1}{h}\phi(\bq') = \frac{1}{h} \phi(\bq + h\bv') \approx \frac{1}{h}\phi(\bq) + {\bf n}\bv' \ge 0. \end{align*} The objective is even cleaner/more intuitive if we denote the next step velocity that would have occurred before the contact impulse is applied as $\bv^- = \bv + h\bM^{-1}\btau$: \begin{align*} \min_{\bv'} \quad & \frac{1}{2} \left(\bv' - \bv^-\right)^T \bM \left(\bv' - \bv^-\right) \\ \subjto \quad & \frac{1}{h}\phi(\bq') \ge 0. \end{align*} The LCP for the frictionless contact dynamics is precisely the optimality conditions of this convex (because $\bM \succeq 0$) quadratic program, and once again the contact impulse, ${\bf f} \ge 0$, plays the role of the Lagrange multiplier (with units $N\cdot s$).

So why do we talk so much about LCPs instead of QPs? Well LCPs can also represent a wider class of problems, which is what we arrive at with the standard transcription of Coulomb friction. In the limit of infinite friction, then we could add an additional constraint that the tangential velocity at each contact point was equal to zero (but these equation may not always have a feasible solution). Once we admit limits on the magnitude of the friction forces, the non-convexity of the disjunctive form rear's it's ugly head.

# Time-stepping LCP: Coulomb Friction

We can use LCP to find a feasible solution with Coulomb friction, but it requires some gymnastics with slack variables to make it work. For this case in particular, I believe a very simple example is best. Let's take our brick and remove the wall and the wheels (so we now have friction with the ground).

The dynamics are the same as our previous example, $$m\ddot{q} = u + f,$$ but this time I'm using $f$ for the friction force which is inside the friction cone if $\dot{q} = 0$ and on the boundary of the friction cone if $\dot{q} \ne 0;$ this is known as the principle of maximum dissipationStewart96. Here the magnitude of the normal force is always $mg$, so we have $|f| \le \mu mg,$ where $\mu$ is the coefficient of friction. And we will use the same semi-implicit Euler approximation to cast this into discrete time.

Now, to write the friction constraints as an LCP, we must introduce some slack variables. First, we'll break the force into a positive component and a negative component: $f[n] = f^+ - f^-.$ And we will introduce one more variable $v_{abs}$ which will be non-zero if the velocity is non-zero (in either direction). Now we can write the LCP: \begin{align*} \find_{v_{abs}, f^+, f^-} \quad \subjto && \\ 0 \le v_{abs} + v[n+1] \quad &\perp& f^+ \ge 0, \\ 0 \le v_{abs} - v[n+1] \quad &\perp& f^- \ge 0, \\ 0 \le \mu mg - f^+ - f^- \quad &\perp& v_{abs} \ge 0,\end{align*} where each instance of $v[n+1]$ we write out with $$v[n+1] = v[n] + \frac{h}{m}(u + f^+ - f^-).$$ It's enough to make your head spin! But if you work it out, all of the constraints we want are there. For example, for $v[n+1] > 0$, then we must have $f^+=0$, $v_{abs} = v[n+1]$, and $f^- = \mu mg$. When $v[n+1] = 0$, we can have $v_{abs} = 0$, $f^+ - f^- \le \mu mg$, and those forces must add up to make $v[n+1] = 0$.

We can put it all together and write an LCP with both normal forces and friction forces, related by Coulomb friction (using a polyhedral approximation of the friction cone in 3d)Stewart96. Although the solution to any LCP can also be represented as the solution to a QP, the QP for this problem is non-convex.

# Anitescu's convex formulation

In Anitescu06, Anitescu described the natural convex formulation of this problem by dropping the maximum dissipation inequalities and allowing any force inside the friction cone. Consider the following optimization: \begin{align*} \min_{\bv'} \quad & \frac{1}{2} \left( \bv' - \bv^-\right)^T \bM \left(\bv' - \bv^-\right) \\ \subjto \quad & \frac{1}{h}\phi(\bq') + \mu {\bf d}_i \bv' \ge 0, \qquad \forall i \in 1,...,m \end{align*} where ${\bf d}_i, \forall i\in 1,...,m$ are a set of tangent row vectors in joint coordinates parameterizing the polyhedral approximation of the friction cone (as in Stewart96). Importantly, for each ${\bf d}_i$ we must also have $-{\bf d}_i$ in the set. By writing the Lagrangian, $$L(\bv', {\bf \beta}) = \frac{1}{2}(\bv' - \bv^-)^T\bM(\bv' - \bv^-) - \sum_i \beta_i \left(\frac{1}{h}\phi(\bq) + ({\bf n} + \mu {\bf d})\bv'\right),$$ and checking the stationarity condition: $$\pd{L}{\bv'}^T = \bM(\bv' - \bv) - h\btau - \sum_i \beta_i ({\bf n} + \mu{\bf d}_i)^T = 0,$$ we can see that the Lagrange multiplier, $\beta_i \ge 0$, associated with the $i$th constraint is the magnitude of the impulse (with units $N \cdot s$) in the direction ${\bf n} - \mu {\bf d}_i,$ where ${\bf n} = \pd{\phi}{\bq}$ is the contact normal; the sum of the forces is an affine combination of these extreme rays of the polyhedral friction cone. As written, the optimization is a QP.

But be careful! Although the primal solution was convex, and the dual problems are always convex, the objective here can be positive semi-definite. This isn't a mistake -- Anitescu06 describes a few simple examples where the solution to $\bv'$ is unique, but the impulses that produce it are not (think of a table with four legs).

When the tangential velocity is zero, this constraint is tight; for sliding contact the relaxation effectively causes the contact to "hydroplane" up out of contact, because $\phi(\bq') \ge h\mu {\bf d}_i\bv'.$ It seems like a quite reasonable approximation to me, especially for small $h$!

Let's continue on and write the dual program. To make the notation a little better, the us stack the Jacobian vectors into $\bJ_\beta$, such that the $i$th row is ${\bf n} + \mu{\bf d_i}$, so that we have $$\pd{L}{\bv'}^T = \bM(\bv' - \bv) - h\btau - \bJ_\beta^T \beta = 0.$$ Substituting this back into the Lagrangian give us the dual program: $$\min_{\beta \ge 0} \frac{1}{2} \beta^T \bJ_\beta \bM^{-1} \bJ_\beta^T \beta + \frac{1}{h}\phi(\bq) \sum_i \beta_i.$$

One final note: I've written just one contact point here to simplify the notation, but of course we repeat the constraint(s) for each contact point; Anitescu06 studies the typical case of only including potential contact pairs with $\phi(\bq) \le \epsilon$, for small $\epsilon \ge 0$.

# Todorov's convex formulation

According to Todorov14, the popular MuJoCo simulator uses a slight variation of this convex relaxation, with the optimization: \begin{align*} \min_\lambda \quad & \frac{1}{2} \lambda^T \bJ \bM^{-1} \bJ^T \lambda + \frac{1}{h}\lambda^T \left( \bJ \bv^- - \dot\bx^d \right), \\ \subjto \quad & \lambda \in \mathcal{FC}(\bq),\end{align*} where $\dot\bx^d$ is a "desired contact velocity", and $\mathcal{FC}(\bq)$ describes the friction cone. The friction cone constraint looks new but is not; observe that $\bJ^T \lambda = \bJ_\beta^T \beta$, where $\beta \ge 0$ is a clever parameterization of the friction cone in the polyhedral case. The bigger difference is in the linear term: Todorov14 proposed $\dot\bx^d = \bJ \bv - h\mathcal{B} \bJ \bv - h \mathcal{K} [\phi(\bq), 0, 0]^T,$ with $\mathcal{B}$ and $\mathcal{K}$ stabilization gains that are chosen to yield a critically damped response.

How should we interpret this? If you expand the linear terms, you will see that this is almost exactly the dual form we get from the position equality constraints formulation, including the Baumgarte stabilization. It's interesting to think about the implications of this formulation -- like the Anitescu relaxation, it is possible to get some "force at a distance" because we have not in any way expressed that $\lambda = 0$ when $\phi(\bq) > 0.$ In Anitescu, it happened in a velocity-dependent boundary layer; here it could happen if you are "coming in hot" -- moving towards contact with enough relative velocity that the stabilization terms want to slow you down.

In dual form, it is natural to consider the full conic description of the friction cone: $$\mathcal{FC}(\bq) = \left\{{\bf \lambda} = [f_n, f_{t1}, f_{t2}]^T \middle| f_n \ge 0, \sqrt{f^2_{t1}+f^2_{t2}} \le \mu f_n \right\}.$$ The resulting dual optimization is has a quadratic objective and second-order cone constraints (SOCP).

There are a number of other important relaxations that happen in MuJoCo. To address the positive indefiniteness in the objective, Todorov14 relaxes the dual objective by adding a small term to the diagonal. This guarantees convexity, but the convex optimization can still be poorly conditioned. The stabilization terms in the objective are another form of relaxation, and Todorov14 also implements adding additional stabilization to the inertial matrix, called the "implicit damping inertia". These relaxations can dramatically improve simulation speed, both by making each convex optimization faster to solve, and by allowing the simulator to take larger integration time steps. MuJoCo boasts a special-purpose solver that can simulate complex scenes at seriously impressive speeds -- often orders of magnitude faster than real time. But these relaxations can also be abused -- it's unfortunately quite common to see physically absurd MuJoCo simulations, especially when researchers who don't care much about the physics start changing simulation parameters to optimize their learning curves!

# Relaxed complementarity-free formulation: Normal force

What equations do we get if we simulate the simple cart colliding with a wall example that we used above?

Following Todorov14, the first relaxation we find is the addition of "implicit damping inertia", which effectively adds some inertia and some damping, parameterized by the regularization term $b$, to the explicit form of the equations. As a result our discrete-time cart dynamics are given by $$(m+b)\frac{v[n+1] - v[n]}{h} = u[n] + f[n] - bv[n].$$

What remains is to compute $f[n]$, and here is where things get interesting. Motivated by the idea of "softening the Gauss principle" (e.g. the principle of least constraint Udwadia92) by replacing the hard constraints with a soft penalty term, we end up with a convex optimization which, in this scalar frictionless case, reduces to: \begin{gather*}f[n] = \argmin_{f \ge 0} \frac{1}{2} f^2 \frac{1 + \epsilon}{m+b} + f(v^- - v^*),\end{gather*} where $v^- = v[n] + \frac{h}{m + b}(u[n] - bv[n])$ is the next-step velocity before the contact impulse. $v^*$ is another regularization term: the "desired contact velocity" which is implemented like a by Baumgarte stabilization: $$v^* = v[n] - 2h \frac{1 + \epsilon}{\kappa} v[n] - h\frac{1+\epsilon}{\kappa^2} q[n].$$ This introduced two additional relaxation parameters; MuJoCo calls $\epsilon$ the "impulse regularization scaling" and $\kappa$ the "error-reduction time constant".

# Beyond Point Contact

Coming soon... Box-on-plane example. Multi-contact.

Also a single point force cannot capture effects like torsional friction, and performs badly in some very simple cases (imaging a box sitting on a table). As a result, many of the most effective algorithms for contact restrict themselves to very limited/simplified geometry. For instance, one can place "point contacts" (zero-radius spheres) in the four corners of a robot's foot; in this case adding forces at exactly those four points as they penetrate some other body/mesh can give more consistent contact force locations/directions. A surprising number of simulators, especially for legged robots, do this.

In practice, most collision detection algorithms will return a list of "collision point pairs" given the list of bodies (with one point pair per pair of bodies in collision, or more if they are using the aforementioned "multi-contact" heuristics), and our contact force model simply attaches springs between these pairs. Given a point-pair, $p_A$ and $p_B$, both in world coordinates, ...

Hydroelastic model in drakeElandt19.

numerical integration. discrete-time approximations of continuous systems (lots of edx people had trouble with this. some found: http://web.mit.edu/10.001/Web/Course_Notes/Differential_Equations_Notes/node3.html)

# Parameter Estimation

lumped parameters. identifiable lumped parameters. least-squares estimation. power formulations. trajectory design.