Within the earlier put up, we launched activity planning in robotics. This area broadly entails a set of planning methods which are domain-independent: That’s, we will design a website which describes the world at some (usually excessive) stage of abstraction, utilizing some modeling language like PDDL. Nonetheless, the planning algorithms themselves may be utilized to any area that may be modeled in that language, and critically, to unravel any legitimate drawback specification inside that area.

The important thing takeaway of the final put up was that activity planning is finally search. These search issues are sometimes difficult and develop exponentially with the scale of the issue, so it’s no shock that activity planning is commonly symbolic: There are comparatively few attainable actions to select from, with a comparatively small set of finite parameters. In any other case, search is prohibitively costly even within the face of intelligent algorithms and heuristics.

Bridging the hole between this summary planning area and the true world, which we deeply care about in robotics, is difficult. In our instance of a cellular robotic navigating a family atmosphere, this would possibly look as follows: Provided that we all know two rooms are linked, a plan that takes our robotic from room A to room B is assured to execute efficiently. In fact, this isn’t essentially true. We’d give you a superbly legitimate summary plan, however then fail to generate a dynamically possible movement plan via a slender hallway, or fail to execute a superbly legitimate movement plan on our actual robotic {hardware}.

That is the place Activity and Movement Planning (TAMP) is available in. What if our planner spends extra effort deliberating about extra concrete features of a plan earlier than executing it? This presents a key tradeoff in additional up-front computation, however a decrease danger of failing at runtime. On this put up we’ll discover just a few issues that differentiate TAMP from “plain” activity planning, and dive into some detailed examples with the pyrobosim and PDDLStream software program instruments.

Some motivating examples

Earlier than we formalize TAMP additional, let’s think about some difficult examples you would possibly encounter with purely symbolic planning in robotics purposes.

On this first instance, our aim is to select up an apple. In purely symbolic planning the place all actions have the identical value, there isn’t a distinction in navigating to table0 and table1, each of which have apples. Nonetheless, you’ll discover that table0 is in an unreachable location. Moreover, if we determine to embed navigation actions with a heuristic value equivalent to straight-line distance from the start line, our heuristic under will favor table0 as a result of it’s nearer to the robotic’s beginning place than table1.

It wouldn’t be till we attempt to refine our plan — for instance, utilizing a movement planner to seek for a sound path to table0 within the unreachable room — that we might fail, must replace our planning area to someway flag that main_room and unreachable are disconnected, after which replan with this new data.

Pathological activity planning instance for goal-directed navigation.
Each table0 and table1 can result in fixing the aim specification of holding an apple, however table0 is totally unreachable.

On this second instance, we wish to place a banana on a desk. As with the earlier instance, we may select to put this object on both desk0 or desk1. Nonetheless, within the absence of extra data — and particularly if we proceed to deal with close by places as decrease value — we could plan to put banana0 on desk0 and fail to execute at runtime due to the opposite obstacles.

Right here, some different options would come with inserting banana0 on desk1, or shifting one of many different objects (water0 or water1) out of the way in which to allow banana0 to suit on desk0. Both means, we want some notion of collision checking to allow our planner to remove the seemingly optimum, however in follow infeasible, plan of merely inserting the thing on desk0.

Pathological activity planning instance for object manipulation.
Inserting banana0 on both desk0 or desk1 will fulfill the aim, however desk0 has different objects that might result in collisions. So, banana0 should both positioned on desk1, or the objects should be rearranged and/or moved elsewhere to permit banana0 to suit on desk0.

In each instances, what we’re lacking from our purely symbolic activity planner is the flexibility to contemplate the feasibility of summary actions earlier than spitting out a plan and hoping for the perfect. Particularly for embodied brokers equivalent to robots, which transfer within the bodily world, symbolic plans should be made concrete via movement planning. As seen in these two examples, what if our activity planner additionally required the existence of a particular path to maneuver between two places, or a particular pose for putting objects in a cluttered area?

What’s Activity and Movement Planning?

In our examples, the core subject is that if our activity planning area is just too summary, a seemingly legitimate plan is prone to fail down the road after we name a very decoupled movement planner to attempt execute some portion of that plan. So, activity and movement planning is precisely as its title states — collectively fascinated with duties and movement in a single planner. As Garrett et al. put it of their 2020 survey paper, “TAMP really lies between discrete “high-level” activity planning and steady “low-level” movement planning”.

Nonetheless, there’s no free lunch. When contemplating all of the fantastic particulars up entrance in deliberative planning, search turns into costly in a short time. In symbolic planning, an motion could have a discrete, finite checklist of attainable targets (let’s say someplace round 5-10), so it might be cheap to exhaustively search over these and discover the one parameter that’s optimum in accordance with our mannequin. Once we begin fascinated with detailed movement plans which have a steady parameter area spanning infinite attainable options, this turns into intractable. So, a number of approaches to TAMP will apply sampling-based methods to make planning work in steady motion areas.

One other means to make sure TAMP is sensible is to leverage hierarchy. One standard approach for breaking down symbolic planning into manageable items is Hierarchical Activity Networks (HTNs). In these 2012 lecture slides, Nilufer Onder mentions “It could be a waste of time to assemble plans from particular person operators. Utilizing the built-in hierarchy helps escape from exponential explosion.” An instance of hierarchical planning is proven within the diagram under. Utilizing this diagram, you’ll be able to discover the advantages of hierarchy; for instance, this planner would by no means must even think about find out how to open a door if the summary plan didn’t require taking place the hallway.

An instance of hierarchical planning for a robotic, the place high-level, or summary, plans for a robotic may very well be refined into lower-level, or concrete, actions.
Supply: Automated Planning and Performing (2016)

Hierarchical planning is nice in that it helps prune infeasible plans earlier than spending time producing detailed, low-level plans. Nonetheless, on this area the legendary downward refinement property is commonly cited. To straight quote the 1991 paper by Bacchus and Yang, this property states that “given {that a} concrete-level resolution exists, each summary resolution may be refined to a concrete-level resolution with out backtracking throughout summary ranges”. This isn’t all the time (and I’d argue hardly ever) achievable in robotics, so backtracking in hierarchical planning is essentially unavoidable.

To this finish, one other technique behind TAMP has to do with dedication in sampling parameters throughout search. Within the literature, you will notice many equal phrases thrown round, however I discover the principle distinction is between the next methods:

  • Early-commitment (or binding) methods will pattern motion parameters from steady area earlier than search, successfully changing the issue to a purely discrete activity planning drawback.
  • Least-commitment (or optimistic) methods will as a substitute give you a purely symbolic plan skeleton. If that skeleton is possible, then the required parameter placeholders are crammed by sampling.

Flowcharts representing two excessive sorts of sampling-based TAMP.
*H-CSP = hybrid constraint satisfaction drawback
Supply: Garrett et al. (2020), Built-in Activity and Movement Planning

Each methods have benefits and drawbacks, and in follow fashionable TAMP strategies will mix them in a roundabout way that works for the varieties of planning domains and issues being thought-about. Additionally, notice that within the diagram above each methods have a loop again to the start when an answer isn’t discovered; so backtracking stays an unavoidable a part of planning.

One key paper that demonstrated the steadiness of symbolic search and sampling was Sampling-based Movement and Symbolic Motion Planner (SMAP) by Plaku and Hager in 2010. Across the identical time, in 2011, Leslie Kaelbling and Tomás Lozano-Pérez offered Hierarchical Planning within the Now (HPN), which mixed hierarchy and sampling-based methods for TAMP. Nonetheless, the authors themselves admitted the sampling half left one thing to be desired. There’s a nice quote on this paper which foreshadows a few of the different work that might come out of their lab:

“As a result of our domains are infinite, we can’t think about all instantiations of the operations. Our present implementation of suggesters solely considers a small variety of attainable instantiations of the operations. We may get better the comparatively weak properties of probabilistic completeness by having the suggesters be mills of an infinite stream of samples, and managing the search as a non-deterministic program over these streams.”

– Leslie pack kaelbling and Tomás Lozano-Pérez (2011), Hierarchical planning within the now.

Straight following this quote is the work their pupil Caelan Garrett took on — first within the creation of STRIPStream in 2017 after which PDDLStream in 2018. The astute reader could have observed that PDDLStream is the precise software program utilized in these weblog posts, so take this “literature evaluation” with this bias in thoughts, and hold studying if you wish to study extra about TAMP with this particular software.

If you wish to know extra about TAMP on the whole, I’ll refer you to 2 current survey papers that I discovered helpful:

Cell robotic instance, revisited

For instance the advantages of built-in TAMP, we’ll proceed the identical cellular robotics instance from the earlier put up. On this drawback,

  • The robotic’s aim is to put the apple on the desk.
  • Navigation now requires developing with a aim pose (which is a steady parameter), as properly the precise path from begin to aim. For this instance, we’re utilizing a Quickly-exploring Random Tree (RRT), however you possibly can swap for some other path-finding algorithm.
  • Inserting an object now requires sampling a sound pose that’s inside the position floor polygon and doesn’t collide with different objects on that floor.

As you learn the next checklist explaining this drawback, be sure to scroll via the slideshow under to get a visible illustration.

STEP 1: Wanting on the state of the world, you’ll be able to see how a purely symbolic activity planner would output a comparatively easy plan: choose the apple, transfer to the desk, and place the apple on the desk. Within the context of TAMP, this now represents a plan skeleton which a number of parameters which are but to be crammed — particularly,

  • ?pt is the pose of the robotic when navigating to the desk
  • ?path is the precise output of our movement planner to get to ?pt
  • ?pa-1 is the brand new pose of the apple when positioned on the desk (which follows from its preliminary pose ?pa-0)

: To make the issue somewhat easier, we made it such that each location has a discrete, finite set of attainable navigation places equivalent to the sides of its polygon. So trying on the desk location, you see there are 4 attainable navigation poses pt-T, pt-B, pt-L, and pt-R equivalent to the highest, backside, left, and proper sides, respectively. Since this set of places is comparatively small, we will pattern these parameters up entrance (or eagerly) firstly of search.

STEP 3: Our transfer motion can now have completely different instantiations for the aim pose ?pt which are enumerated throughout search. That is in distinction with the ?path argument, which should be sampled by calling our RRT planner. We don’t wish to do that eagerly as a result of the area of paths is steady, so we favor to defer sampling of this parameter. If our motion has a value related to the size of a path, we may think about that the lowest-cost motion could be to navigate to the left aspect of the desk (pt-L), and a few randomly sampled path (path42) could describe how we get there.

STEP 4: Subsequent comes the place motion, which now should embrace a sound collision-free pose for the apple on the desk. Due to how we arrange our drawback, our robotic can’t discover a legitimate placement pose when approaching from the left aspect of the desk. So, we should backtrack.

STEP 5: After backtracking, we have to discover an alternate navigation pose for the desk (?pt). Given our surroundings, the one different possible location is the underside aspect of the desk (pt-b), because the partitions block the robotic from the highest and proper sides and it could be unimaginable to discover a legitimate path with our RRT. Nonetheless, when the robotic is on the backside aspect of the desk, it could additionally pattern a sound placement pose! In our instance, the placeholder ?pa-1 is subsequently glad with some randomly sampled pose pa29.

STEP 6: … And there you could have it! A legitimate plan that defines a sequence of symbolic actions (choose, transfer, place) together with the required navigation pose, path to that pose, and placement location for the apple. It’s not optimum, however it’s probabilistically full!

(1/6) By being optimistic about all the continual parameters associated to movement, we will attain a possible aim state with relative ease.

(2/6) Because the navigation poses across the desk and the desk are finite, we will pattern them eagerly; that’s, we enumerate all choices up entrance in planning.

(3/6) As soon as we decide to a navigation pose across the desk, we will proceed filling in our plan by sampling a possible trajectory from the robotic’s present pose to the goal pose on the desk.

(4/6) Subsequent, we have to pattern a placement pose for the apple. Suppose on this case we fail to pattern a collision-free resolution based mostly on the robotic’s present location.

(5/6) This implies we have to backtrack and think about a special navigation pose, thereby a special movement plan to this new pose.

(6/6) From this new pose, despite the fact that the trajectory is longer and subsequently higher-cost, we will pattern a sound placement pose for the apple and at last full our activity and movement plan.

Now, suppose we alter our surroundings such that we will solely method the desk from the left aspect, so there isn’t a approach to straight discover a legitimate placement pose for the apple. Utilizing the identical planner, we should always ultimately converge on a activity and movement plan that rearranges the objects world — that’s, it requires shifting one of many different objects on the desk to make room for the apple.

Implementing TAMP with PDDLStream

We are going to now revisit our pathological examples from the start of this put up. To do that, we’ll use PDDLStream for planning and pyrobosim as a easy simulation platform. For fast background on PDDLStream, it’s possible you’ll consult with this video.

The important thing thought behind PDDLStream is that it extends PDDL with a notion of streams (keep in mind the sooner quote from the Hierarchical Planning within the Now paper?). Streams are generic, user-defined Python features that pattern steady parameters equivalent to a sound pattern certifies that stream and supplies any obligatory predicates that (normally) act as preconditions for actions. Additionally, PDDLStream has an adaptive approach that balances exploration (looking for discrete activity plans) vs. exploitation (sampling to fill in steady parameters).

Aim-directed navigation

We will use PDDLStream to enhance our transfer motion such that it contains metric particulars in regards to the world. As we noticed in our illustrative instance, we now should issue within the begin and aim pose of our robotic, in addition to a concrete path between these poses.

As extra preconditions for this motion, we should make sure that:

  • The navigation pose is legitimate given the goal location (NavPose)
  • There should be a sound path from the begin to aim pose (Movement)

Moreover, we’re in a position to now use extra real looking prices for our motion by calculating the precise size of our path produced by the RRT! The separate file describing the streams for this motion could look as follows. Right here, the s-navpose stream certifies the NavPose predicate and the s-motion stream certifies the Movement predicate.

The Python implementations for these features would then look one thing like this. Discover that the get_nav_poses perform returns a finite set of poses, so the output is a straightforward Python checklist. Then again, sample_motion can repeatedly spit out paths from our RRT, and it carried out as a generator:

Placing this new area and streams collectively, we will clear up our first pathological instance from the introduction. Within the plan under, the robotic will compute a path to the farther away, however reachable room to select up an apple and fulfill the aim.

Object manipulation

Equally, we will prolong our place motion to now embrace the precise poses of objects on this planet. Particularly,

  • The ?placepose argument defines the goal pose of the thing.
  • The Placeable predicate is licensed by a s-place stream.
  • The IsCollisionFree predicate is definitely a derived predicate that checks particular person collisions between the goal object and all different objects at that location.
  • Every particular person collision examine is set by the CollisionFree predicate, which is licensed by a t-collision-free steam.

The Python implementation for sampling placement poses and checking for collisions could look as follows. Right here, sample_place_pose is our generator for placement poses, whereas test_collision_free is a straightforward Boolean (true/false) examine.

By increasing our area to motive in regards to the feasibility of object placement, we will equally clear up the second pathological instance from the introduction. Within the first video, we’ve an alternate location desk1 the place the robotic can place the banana and fulfill the aim.

Within the second video, we take away the choice desk1. The identical activity and movement planner then produces an answer that entails selecting up one of many objects on desk0 to make room to later place the banana there.

You possibly can think about extending this to a extra real looking system — that’s, one that isn’t some extent robotic and has an precise manipulator — that equally checks the feasibility of a movement plan for selecting and inserting objects. Whereas it wasn’t the principle focus of the work, our Lively Studying of Summary Plan Feasibility work did precisely this with PDDLStream. Particularly, we used RRTs to pattern configuration-space paths for a Franka Emika Panda arm and doing collision-checking utilizing a surrogate mannequin in PyBullet!


On this put up we launched the final idea of activity and movement planning (TAMP). In idea, it’s nice to deliberate extra — that’s, actually suppose extra in regards to the feasibility of plans right down to the metric stage — however with that comes extra planning complexity. Nonetheless, this could repay in that it reduces the chance of failing in the course of executing a plan and having to cease and replan.

We launched 3 basic ideas that may make TAMP work in follow:

  • Hierarchy, to find out the feasibility of summary plans earlier than planning at a decrease stage of refinement.
  • Steady parameter areas, and methods like sampling to make this tractable.
  • Least-commitment methods, to give you symbolic plan skeletons earlier than spending time with costly sampling of parameters.

We then dug into PDDLStream as one software for TAMP, which doesn’t do a lot in the way in which of hierarchy, however definitely tackles steady parameter areas and least-commitment methods for parameter binding. We went via just a few examples utilizing pyrobosim, however you’ll be able to entry the total set of examples within the pyrobosim documentation for TAMP.

The PDDLStream repository has many extra examples that you could try. And, in fact, there are various different activity and movement planners on the market that target various things — equivalent to hierarchy with out steady parameters, or factoring in different frequent targets equivalent to temporal features and useful resource consumption.

Hope you could have loved these posts! If the instruments proven right here offer you any cool concepts, I’d love to listen to about them, so be at liberty to achieve out.

You possibly can learn the unique article at Roboticseabass.com.

Sebastian Castro
is a Senior Robotics Engineer at PickNik.

By admin

Leave a Reply

Your email address will not be published.