Resolving Conflicts Where and When They Arise: Reactive Composition of Multi-Goal Behavior

arXiv:2605.27314 · cs.RO, cs.SY, eess.SY · Submitted 2026-05-26 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Robotics Radio. Generated commentary on the latest robotics and control papers.

Rosa: Today's paper: "Resolving Conflicts Where and When They Arise".

Dev: Reactive control is often considered insufficient for multi-objective tasks because conflicting objectives give rise to local minima.

Rosa: First, who's behind it and why it matters.

Title and authors: Rosa: So we're looking at this paper titled "Resolving Conflicts Where and When They Arise: Reactive Composition of Multi-Goal Behavior," which seems to tackle a really fundamental problem where simple reactive control just doesn't cut it for tasks with multiple competing objectives because they can lead to local minima.

Dev: Yeah, that's exactly what caught my attention; the title suggests they are looking at *when* and *where* those conflicts happen and how to handle them reactively instead of relying on some static planning or fixed rules.

Taro: I'm curious about the authors, who are Vito Mengers and Oliver Brock, because their approach seems to be based on extending graph-based world models with nullspace projections to resolve these conflicts dynamically based on the current state.

Rosa: It sounds like they aren't just looking at a general problem; they are grounding this in a specific architecture, using AICON, which is this graph-based representation that encodes sensory inputs, actions, and time through state-dependent active interconnections.

Dev: That's the core mechanism we need to focus on for the engineering side; I wonder how their proposed method handles the computational load when they have to compute these nullspace projections in real-time.

Taro: I'm thinking about what happens when things go wrong, like if the world misbehaves unexpectedly; does this dynamic resolution mean better handling of sudden changes compared to a standard steepest descent approach?

Rosa: Well, the paper explains that instead of just picking one steepest gradient, they combine gradients from different paths by projecting lower-priority ones into the nullspace of higher-priority ones based on continuous priorities derived from the current state.

Dev: That projection idea is interesting because it sounds like a way to mathematically disentangle competing forces without having to predefine a rigid hierarchy for all scenarios. It's a dynamic prioritization system.

Taro: So, when the world throws us a wrench in the works, does this system have an exploration mode built in, or does it just stick rigidly to the projected gradients?

Rosa: The paper describes a specific exploration mechanism triggered when conflicts are detected—when the two strongest gradients show a cosine similarity below a threshold of negative zero point six, signaling near opposition.

Title and authors: Dev: That threshold sounds like a critical piece of tuning for the stability of the system; if that threshold is too sensitive, we could get oscillatory behavior instead of finding a solution.

Taro: And when that exploration mode kicks in, how does it decide where to go in that nullspace? Does it have some memory or consistency check built into its movement during those conflict situations?

Rosa: The system moves along the direction within the nullspace of the dominant gradient that is most consistent with recent motion history, which they represent as an exponential moving average of that corresponding quantity.

Dev: That reliance on a moving average for consistency is something I'll want to scrutinize; it sounds like they are using temporal smoothing to keep things from jumping around too much when the objectives are fighting.

Taro: Speaking of those conflicting objectives, the paper shows success in two key areas: navigation around non-convex obstacles and planar pushing of non-convex objects, where they claim one hundred percent success across one hundred configurations against zero percent for a simple steepest descent baseline <ref:2605.27314#pg0,and planar pushing of non-convex objects, where>.

Rosa: That one hundred percent success rate in push tasks is quite compelling; it shows that this method can handle complex physical interactions far better than what we see with standard potential fields or fixed hierarchies <ref:2605.27314#pg0>.

Dev: I'm interested in the transferability aspect; they mentioned that the same formulation transfers directly to a real robot, incorporating perceptual and kinematic constraints like joint limits seamlessly. That’s a big deal for deployability.

Taro: If it works on a real robot with those physical limitations, does this mean we can finally push autonomous agents into more physically complex environments where static encodings fail completely?

Rosa: It seems the authors argue that the limitation isn't just in the control, but in those static encodings themselves, suggesting that online negotiation between objectives can actually replace traditional planning in complex tasks.

Dev: That shifts the burden away from needing perfect offline planning for every possible scenario; instead, we rely on this online negotiation mechanism to manage dynamics.

Title and authors: Taro: Thinking about the big picture implications, if this approach generalizes well across different domains and handles those structural conflicts better than previous methods like diffusion policies, what does that suggest for future autonomy?

Rosa: It implies that a more robust way to handle multi-objective control exists where priorities are continuously adjusted based on how things are currently interacting in the state space.

Dev: For the control loop rate, I need to know if this nullspace projection calculation adds significant latency; if it does, we might have trouble maintaining the stability they claim.

Taro: If they can successfully resolve conflicts like pushing an object against an obstacle, that has massive implications for human-robot collaboration and navigating cluttered spaces autonomously.

Rosa: It really does suggest that we need to move away from fixed hierarchies in control systems and toward mechanisms that adapt their priorities based on real-time interaction forces.

Dev: So, the paper’s main contribution seems to be providing a mathematically sound way to achieve this dynamic conflict resolution using nullspace projections within a graph model.

Taro: I'm also impressed by how they showed invariance to absolute goal pose and actuation speed constraints, which is something that makes it much more practical for real-world deployment.

Rosa: It really shows that the non-stationarity we observe in complex problems might stem from the structure of the problem itself, not just random noise injected into the system.

Dev: That’s a strong point; if we can use structural conflict resolution to escape unproductive dynamics, it means our control systems don't have to constantly fight against unpredictable disturbances.

Taro: So, to wrap up this discussion on "Resolving Conflicts Where and When They Arise: Reactive Composition of Multi-Goal Behavior," the paper offers a way to move past static encodings by using continuous nullspace projections to negotiate priorities dynamically.

Rosa: It’s a really interesting piece of work that moves control theory into a more adaptive regime for multi-objective tasks, and I'm eager to see how this translates beyond simulation.

Dev: I just hope the real-time implementation is as stable as they show, especially concerning those projection calculations we discussed earlier.

Taro: Overall, the implication is that we can build agents that are much more resilient when faced with simultaneous, conflicting demands in dynamic environments like navigation and manipulation.

The paper's summary: Rosa: So, essentially, this paper argues that traditional reactive control falls short for multi-objective tasks because conflicting goals lead to those nasty local minima we always run into in practice, and they propose extending graph models with nullspace projections to dynamically resolve those conflicts based on how priorities change as the system evolves.

Dev: That extension sounds mathematically intense; I'm thinking about the practical side—how does this dynamic prioritization actually translate into stable control signals without introducing massive latency that would kill our loop rate?

Taro: From an autonomy standpoint, I'm interested in what happens when the world throws us a wrench in the works; does this nullspace negotiation give the AI enough flexibility to pivot effectively instead of just getting stuck following a single, failing gradient?

Rosa: Exactly; they show that by projecting lower-priority gradients into the nullspace of higher ones, you can manage those overlaps where objectives fight each other without needing a pre-programmed hierarchy that breaks down when priorities shift.

Dev: But how do they define those continuous priorities in real time? If the priority ordering isn't fixed beforehand, what’s the mechanism that decides which gradient is "higher" right now?

Taro: They determine this continuously from the current gradient magnitudes; it’s a dynamic system where the system figures out its own relative importance as it moves through state space.

Rosa: And when they detect that two strong gradients are actually in opposition, they enter an exploration mode that uses the nullspace of the dominant goal to search for a path consistent with recent movement history, which is pretty smart.

Dev: That sounds like a sophisticated way to handle those critical conflict points we discussed earlier; I worry about the stability when it switches between pursuit and this exploration phase.

Taro: The authors claim success in complex scenarios like pushing non-convex objects and navigating around obstacles, achieving one hundred percent success in simulations where simpler methods fail, which suggests a much more robust way for AI to interact physically with the world.

Rosa: That one hundred percent success rate is really telling; it moves the problem from an edge case failure to a general capability of online negotiation between competing objectives.

Dev: If this framework can handle physical constraints and perceptual uncertainty simultaneously, that opens up possibilities for real robots that are currently too complex for static planning approaches.

Taro: The implication is that we might stop needing perfect offline plans and instead rely on this continuous, online negotiation mechanism to manage the dynamics of a complex physical system.

Rosa: It really suggests that the landscape of optimization isn't just shaped by injected noise, but by the structure of the problem itself, and this method lets us navigate those structural conflicts directly.

Dev: I’m still focused on implementation; if we can get this stability right in simulation, the next challenge is making sure that calculation doesn't lag behind the physical reality of a moving robot.

Taro: The future work they mention seems to be about ensuring this mechanism generalizes beyond simple tasks and handles more intricate, high-dimensional problems where those structural conflicts become incredibly dense.

Rosa: It sounds like we are looking at a significant step toward building AI that can truly negotiate with the environment rather than just reacting to it.

The paper's improvements: Rosa: So, the paper doesn't just stop at proposing AICON; they suggest several concrete improvements to make this framework more viable for real-world deployment across different domains like manipulation and navigation.

Dev: I'm keen on those suggestions because a system that only works in a perfect simulation isn't useful to me; specifically, how does the paper address incorporating perceptual constraints and joint limits without needing massive retraining?

Taro: I see the focus on dynamic adaptation is key here; they propose replacing fixed trade-offs with continuously adapting interactions so the system can resolve conflicts as state changes, which addresses that myopic gradient following we talked about.

Rosa: They suggest a shift toward biologically plausible decision-making by leveraging nullspace projection mechanisms similar to how human motor control stabilizes task dimensions while leaving secondary objectives free for exploration, which feels like a really deep conceptual leap.

Dev: That idea of stabilizing the task dimensions while allowing secondary objectives freedom is interesting; it hints at a way to manage complexity without overwhelming the control loop with every possible constraint simultaneously.

Taro: And they suggest creating optimization landscapes that are non-stationary by design, meaning the effective optimization surface shifts continuously as priorities change, which should allow it to escape unproductive dynamics through structural conflict resolution rather than relying only on injected action noise.

Rosa: That moves the focus from just reacting to disturbances toward designing a system whose very structure evolves in response to conflicting demands, which is a significant conceptual change for control theory.

Dev: From an engineering standpoint, I need to understand how those mechanisms—the hysteresis and temporal smoothing they introduced—actually affect the stability margins under high-frequency noise or sudden environmental changes.

Taro: The authors show success in transferring this to a real robot, incorporating uncertainty reduction and kinematic limits seamlessly; that suggests the conflict resolution mechanism itself is robust enough to handle those physical realities without needing separate safety layers for every constraint.

Rosa: It really highlights that the boundary between planning and control isn't just a theoretical line anymore; it reflects how well we’ve encoded our objective representations in the first place.

Dev: If this framework can indeed integrate perception and kinematics into the same gradient-based conflict resolution process, that drastically simplifies the architecture for complex hardware integration.

Taro: The implication is that we might be able to build agents that are much more resilient when faced with simultaneous, conflicting demands in dynamic environments like navigation and manipulation, moving beyond simple reactive policies entirely.

Rosa: It feels like this work points toward a future where control systems don't just execute commands but actively negotiate the trade-offs between those commands as they happen.

Conclusion: Rosa: So to wrap up, this paper on "Resolving Conflicts Where and When They Arise: Reactive Composition of Multi-Goal Behavior" essentially shows us how to move beyond static goal hierarchies by using nullspace projections to let competing objectives negotiate their priorities dynamically based on the current state of the system.

Dev: That dynamic negotiation is what makes me hopeful about its potential, but I still have those concerns about latency and whether that continuous adjustment holds up under high-frequency disturbances in a real loop.

Taro: I'm glad they showed how it handles those misbehavior scenarios; when the world throws us a wrench in the works, this system seems designed to pivot rather than just lock onto a single, failing gradient.

Rosa: It really does suggest that we might be able to build AI that can truly negotiate with the environment instead of just reacting to it blindly.

Dev: If we can get this stability right in simulation, the next big hurdle is making sure that calculation doesn't introduce enough lag to destabilize a fast-moving robot.

Taro: The authors’ success in transferring this to physical robots with perceptual constraints shows that the core mechanism is robust enough to handle those real-world limitations without needing separate safety layers for every single constraint.

Rosa: It feels like this work points toward a future where control systems don't just execute commands but actively negotiate the trade-offs between those commands as they happen.

Dev: I'm still focused on the hardware; how do we actually implement these projection calculations efficiently enough to keep up with real-time demands?

Taro: The implication is that we might be able to build agents that are much more resilient when faced with simultaneous, conflicting demands in dynamic environments like navigation and manipulation.

Rosa: Overall, this paper on "Resolving Conflicts Where and When They Arise: Reactive Composition of Multi-Goal Behavior" provides a solid mathematical foundation for achieving more adaptive multi-objective control.

Dev: I just hope we can get this stability right in simulation before we start thinking about deployment on physical hardware.

Taro: I think the next big step for this line of research will be to see how these mechanisms scale up to handle even denser, more intricate problem structures where those conflicts become incredibly complex.

Robotics and Biology Laboratory, Technische Universität Berlin · Science of Intelligence (SCIoI), Cluster of Excellence, Berlin, Germany · Robotics Institute Germany

cs.RO, cs.SY, eess.SY

Submitted: 2026-05-26

Updated: 2026-10-07

License: http://creativecommons.org/licenses/by-sa/4.0/

Importance score: 88/100

The gist: Reactive control is often considered insufficient for multi-objective tasks because conflicting objectives give rise to local minima.

Key concepts

Active InterCONnect (AICON)
A graph-based representation that encodes regularities between sensory inputs, actions, and time. It uses state-dependent active interconnections to model how different parts of the environment interact over time.
Nullspace Projections
A mathematical technique used to resolve conflicts between gradients from different goals. Lower-priority gradients are projected into the nullspace of higher-priority ones, effectively allowing objectives with different priorities to coexist without direct opposition.
Conflict Resolution Mechanism
A dynamic process where the system detects when two strong gradients oppose each other (low cosine similarity). It enters an exploration mode along the dominant gradient's nullspace until conflicts subside or a priority change is justified.
Hysteresis and Smoothing
Mechanisms used for stability. Hysteresis prevents rapid switching between objective priorities by requiring a significant margin before changing order. Temporal smoothing filters the final control signal to ensure stable, continuous motion.

Terminology

Summary

Reactive control is often considered insufficient for multi-objective tasks because conflicting objectives give rise to local minima. This work argues that this limitation arises from static encodings that fail to reflect how objectives currently interact, proposing an extension of graph-based world models with nullspace projections to resolve conflicts dynamically.

The gist

We exploit the interaction structure encoded in a graph-based world model by extending it with nullspace projections: conflicts are resolved where they arise by projecting lower-priority gradients into the nullspace of higher-priority ones, with priorities determined continuously from the current state.

How it works

The framework is built upon Active InterCONnect (AICON), a graph-based representation that encodes regularities between sensory inputs (S), actions (A), and time (t) through state-dependent active interconnections. Goals are expressed as differentiable scalar cost functions, and their gradients are propagated backward through the network of interconnected estimation components using the chain rule to reach actuation signals. This structure allows for sequential behavior by executing the steepest gradient at every time step, but it cannot cope with simultaneous conflicting goals.

How it works (Continued)

To address this limitation, AICON replaces the steepest-selection rule with a combination of gradients through their nullspaces. Specifically, given a gradient along path p1 and a second path p2, the conflict-free combined gradient is defined as: ∇p1▷2 q g = ∇p1 q g + N1 ∇p2 q g. The projector N1 is computed using the pseudoinverse J h-1 of the respective Jacobian, but for scalar cost functions, it simplifies to an orthogonalization step: N1 = I−Q1QT. Crucially, the priority ordering is not fixed in advance but determined continuously from current gradient magnitudes.

How it works (Continued)

Conflict resolution occurs through a layered process where gradients of progressively lower-priority goals are projected into the orthogonal complements of higher-priority ones wherever they begin to overlap. A dynamic selection rule is employed: at each quantity q, we select the steepest gradient in the remaining space as long as degrees of freedom remain, analogously to the single-goal case. This involves selecting ∇[i]q g = argmax ∇g∈G[i]q∥∇g∥, where G[i]q is the set of all candidate gradients for quantity q after previous projections.

How it works (Continued)

Escaping local minima is achieved when conflicts are detected—when the two strongest gradients have a cosine similarity below a threshold θenter = −0.6, indicating near-opposition. Upon detection, the system enters exploration mode: it moves along the direction within the nullspace of the dominant gradient that is most consistent with recent motion history, represented as an exponential moving average of the corresponding quantity. Exploration terminates when cosine similarity rises above θexit = −0.4 or one gradient exceeds all others in magnitude by a factor λ = 3, at which point normal goal pursuit resumes.

How it works (Continued)

To ensure stability and robustness, several mechanisms are introduced:

  1. Hysteresis is applied to priority selection: a change in ordering is only accepted if a competing gradient exceeds the current one by a margin of 10%.

  2. Temporal smoothing is applied to the resulting control signal using a low-pass filter.

  3. Gradient magnitudes are normalized before projection using a softmax with temperature τ = 0.8, which decouples qualitative priority structure from quantitative scaling, improving robustness to reparametrization of cost functions.

How it works (Continued)

The framework demonstrates success in two central domains: navigation around non-convex obstacles and planar pushing of non-convex objects. In the pushT task, AICON successfully resolves conflicts by navigating along the obstacle boundary until a path opens. When transferred to a real robot with perceptual and kinematic constraints, additional goals such as reducing perceptual uncertainty and respecting joint limits are seamlessly incorporated through the same conflict resolution mechanism, showing that the boundary between planning and control reflects the limitations of static objective representations.

How it works (Continued)

Quantitative results across one-hundred randomly generated configurations show that AICON leverages nullspaces consistently resolves conflicts, achieving 100% success under realistic action noise in simulation. The steepest-descent baseline achieves 0% success, with failures arising from local minima at structural conflict points. This demonstrates that online negotiation between objectives can replace planning in complex, multi-objective tasks.

How it works (Continued)

The method is shown to be invariant to absolute goal pose and actuation speed constraints, unlike diffusion policies which degrade significantly under these variations. The results suggest that landscape non-stationarity arises not only from injected randomness but from the structure of the problem itself, framing conflict resolution as a general mechanism for escaping unproductive dynamics.

Improvements for AI systems

Based on the provided paper, here are specific improvements that can be made to AI systems by implementing AICON (Active InterCONnect) with nullspace projections for multi-objective control:

  1. Improve robustness in real-time, reactive robotic manipulation tasks (e.g., object pushing, navigation around non-convex obstacles) by resolving simultaneous conflicting objectives (like goal pursuit vs. obstacle avoidance or beneficial vs. detrimental contacts).

  2. Enable robots to achieve 100% success rates in complex push tasks across various configurations where traditional methods fail, by dynamically negotiating competing forces rather than relying on pre-programmed hierarchies or fixed potential fields.

  3. Allow AI agents to operate effectively in high-dimensional, constrained environments (like real robots) by seamlessly incorporating perceptual constraints (e.g., uncertainty reduction via RGBD vision) and kinematic limits into the same unified gradient-based conflict resolution framework without requiring extensive model retraining or manual constraint coding.

  4. Enhance the adaptability of learned policies by replacing fixed trade-offs with continuously adapting interactions, allowing the system to resolve conflicts online as state changes, thus preventing local minima that trap myopic gradient following.

  5. Develop AI systems that exhibit biologically plausible decision-making by leveraging nullspace projection mechanisms similar to those found in human motor control (stabilizing task dimensions while leaving secondary objectives free for exploration), potentially leading to more robust and energy-efficient behaviors.

  6. Create optimization landscapes that are non-stationary by design; the system's effective optimization surface shifts continuously as priorities change, allowing it to escape unproductive dynamics through structural conflict resolution rather than relying solely on injected action noise (which only resolves periodicity, not structural conflicts).

In summary, the improved AI systems can perform complex, multi-goal tasks reactively—such as navigating cluttered spaces while simultaneously pushing objects—by dynamically negotiating conflicting objectives in real-time using gradient projection into nullspaces.

Abstract

Multi-goal robotic tasks are commonly delegated to planning, because reactive control is prone to local minima when objectives conflict. We show that many such failures stem from static goal representations, not task complexity: the designer defines the objectives, but where and when they must be traded off follows from the world's interaction structure which changes with the state. To read this structure, we extend Active InterCONnect (AICON), a graph of recursive estimators coupled by differentiable interconnections, with adaptive nullspace projections: lower-priority gradients enter the nullspace of higher-priority ones wherever they meet along the graph, not only in action space, ordered by current gradient magnitudes, not a fixed hierarchy. Where two gradients oppose and no weighting of them makes progress, the system instead explores in the nullspace of the stronger one until the conflict dissolves. We robustly solve 100 non-convex navigation and 100 pushT problems, outperforming static potential fields, a diffusion policy, and the same method without projection. Unmodified but embedded in a larger graph, it absorbs perceptual uncertainty, joint limits, and self-collisions on a real robot, solving 49 of 50 pushing trials across varied objects, with active camera control and disturbance recovery emerging from the same coupling. Much of the behavior attributed to planning may thus be within reach of control, once goal representations compose from the conflicts they encounter.

Sources

Related papers