{"id":"28d4560c-2e33-41e1-b96f-6657d54ca479","arxiv_id":"2606.09719","paper_version":1,"verdict":"UNVERDICTED","confidence":"LOW","novelty_score":7.0,"correctness_risk":"unknown","formal_verification":"none","parameter_count":1,"one_line_summary":"A polytope containment condition is enforced as discrete-time CBF constraints inside MPC for local planning whose constraint count depends on geometry complexity rather than obstacle count.","lead":"The paper proposes a motion planning method for robots with complex polygonal shapes that keeps the entire robot footprint inside a convex safe region by adding control barrier function constraints to a model predictive controller. This could let robots move through narrow passages more effectively than methods that simplify the robot to a point or circle.","discovery_kind":"new_method","skeptic_critique":{"model":"grok-4.3","headline":"No significant objection identified","rationale":"Reader's weakest assumption matches the only plausible point of fragility (convex free-space representation from raw sensor data). Because the abstract already states the method works on both grids and LiDAR without segmentation, and no counter-example or hidden assumption appears in the claim structure, the UNVERDICTED verdict stands; the low confidence is simply due to abstract-only access rather than any detected flaw.","tokens_in":1720,"tokens_out":321,"duration_ms":19067,"concrete_test":"Extract the exact inequality set used for containment (likely in §III or §IV) and count the number of CBF constraints as a function of free-space facet count N and robot vertex count M; confirm it equals N·M (or equivalent) and remains constant when the number of obstacles is artificially increased while keeping the same local convex free-space polytope.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim—that polytope containment can be encoded as a fixed number of discrete-time CBF inequalities inside an MPC whose count depends only on local free-space facets and robot vertices, independent of obstacle count, and without explicit obstacle segmentation—follows directly from standard polytope-in-polytope containment (each robot vertex satisfying all half-plane inequalities of the free-space polytope) and the usual discrete CBF form h(x_{k+1}) ≥ (1−α)h(x_k). No internal contradiction with the stated assumptions or with the reported 10 Hz real-time performance on occupancy-grid and LiDAR data is visible.","agreement_with_reader":"agree"},"referee_report":{"model":"grok-4.3","summary":"The paper proposes a local motion planning and control framework for mobile robots with polytopic footprints that guarantees containment inside a continuously updated convex free-space polytope. Containment is encoded as a collection of discrete-time CBF inequalities inside an MPC; the number of inequalities scales with the number of free-space facets and robot vertices rather than the number of obstacles, and the formulation requires no explicit obstacle detection or segmentation. Comparative experiments report up to a 91× reduction in computation time relative to a polytope-based obstacle-avoidance baseline, and the method is demonstrated in simulation on an autonomous surface vehicle and on hardware with a non-holonomic robot using occupancy grids and LiDAR, achieving real-time 10 Hz operation on embedded hardware.","tokens_in":1839,"tokens_out":412,"duration_ms":20363,"significance":"If the central claim holds, the work supplies a practical, scalable alternative to conservative point/circle approximations or per-obstacle CBFs for footprint-aware navigation in tight spaces. The explicit grounding in standard polytope containment and discrete CBF theory, together with the reported hardware validation at 10 Hz, constitutes a concrete engineering contribution that could be adopted in real-time robotics pipelines.","major_comments":[],"minor_comments":[{"comment":"Abstract and §1: the 91× computation-time claim is central to the scaling argument; the manuscript should state the exact obstacle counts, baseline implementation details, and timing measurement protocol used to obtain this figure.","section":null},{"comment":"The description of how the free-space polytope is extracted from occupancy grids or LiDAR (mentioned in the abstract) should include a brief algorithmic outline or reference so that the “no segmentation” claim can be reproduced.","section":null},{"comment":"Notation: the discrete-time CBF class-K function α(·) and the MPC horizon length appear in the formulation but are not given explicit symbols in the abstract; consistent notation would aid readability.","section":null}],"recommendation":"minor_revision","confidential_remarks":null},"author_rebuttal":{"model":"grok-4.3","summary":"We thank the referee for their positive assessment of the manuscript and for recommending minor revision. The provided summary accurately captures the core contribution of encoding polytope containment as discrete-time CBF constraints within MPC, with scaling independent of obstacle count.","responses":[],"tokens_in":1262,"tokens_out":64,"duration_ms":11406,"standing_objections":[]},"desk_editor":{"model":"grok-4.3","letter":"The main takeaway is that this work encodes robot-in-free-space containment as a fixed set of discrete-time CBF inequalities inside an MPC. The number of constraints scales with the number of facets in the local convex free-space polytope and the robot vertices, not with the number of obstacles, and it skips explicit segmentation. That is the concrete novelty relative to standard polytope-obstacle avoidance.\n\nWhat the paper does well is the scaling experiment and the hardware run. The comparative timing shows up to 91x speedup as obstacle count grows, and they close the loop at 10 Hz on an embedded computer using both grid and LiDAR data for an ASV in simulation and a non-holonomic robot on hardware. Those numbers are the kind of evidence that matters for a control paper.\n\nThe soft spots are modest. The central assumption—that the continuously updated free space can be treated as a single convex polytope containing the robot footprint—holds in the tested environments but will need checking in cluttered or non-convex scenes where the free-space representation itself becomes the bottleneck. The abstract mentions comparative analysis and validation, but without the full derivations or raw timing tables in front of me it is hard to judge how tight the CBF class-K functions were tuned or whether the discrete-time formulation introduces noticeable conservatism. No load-bearing circularity or invented entities appear.\n\nThis is the sort of paper a robotics control group would want to see. It is for readers who already work with CBFs or polytope methods and want a concrete, implementable reduction in constraint count. It deserves a serious referee because the scaling claim is falsifiable and the hardware result is reproducible in principle. I would send it out for review.","headline":"The paper gives a workable CBF-MPC encoding for keeping a polytopic robot inside an updated free-space polytope, with constraint count tied to geometry rather than obstacle count.","tokens_in":2306,"tokens_out":420,"would_cite":false,"duration_ms":11187,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"grok-4.3","headline":"A polytopic robot footprint is kept inside a convex free-space region by encoding containment as discrete-time control barrier function constraints in a model predictive controller.","keywords":["motion planning","control barrier functions","model predictive control","polytope containment","robot footprint","free-space representation","autonomous mobile robots","safety constraints"],"falsifier":"A scenario with many obstacles where the computation time does not decrease substantially compared to obstacle-based methods or where the robot violates containment despite the constraints being active.","tokens_in":2626,"feed_emoji":"🤖","tokens_out":637,"duration_ms":14975,"temperature":0.7,"pith_summary":"The paper establishes a motion planning method for robots with physical footprints in tight spaces by representing the free-space as a convex polytope and enforcing containment through control barrier functions. This approach integrates the safety constraints directly into a model predictive controller without needing separate obstacle detection or segmentation. The number of constraints scales with the geometry of the free-space and robot shape rather than the count of obstacles, which improves computational efficiency as obstacle numbers grow. Validation shows real-time performance at 10 Hz on embedded hardware for both simulation and physical robots using occupancy grids and LiDAR.","feed_headline":"Barrier functions enforce robot polytope containment in free space","feed_subtitle":"Safety constraints scale with local geometry complexity instead of obstacle count, supporting real-time planning up to 91 times faster.","key_machinery":"Discrete-time control barrier function (CBF) constraints that encode the polytope-in-polytope containment condition, incorporated into the model predictive controller (MPC).","core_discovery":"The containment condition between the robot's polytopic footprint and the continuously updated convex free-space polytope is formulated as a set of discrete-time control barrier function constraints within a model predictive controller. This guarantees that the robot stays inside the free-space region, and the formulation does not require obstacle detection or segmentation steps.","pith_inferences":["Extending this to non-convex free-spaces might require decomposition into multiple polytopes while retaining the CBF encoding.","This could combine with global path planners for complete navigation stacks in larger environments.","Applications in higher-dimensional spaces like 3D robotic arms could benefit from similar scaling properties if polytopic representations are used."],"forward_implications":["The number of safety constraints depends on free-space geometry complexity and robot shape, not the number of obstacles.","A comparative analysis shows up to 91 times reduction in computation time as the number of obstacles increases.","The method enables safe real-time motion planning and control at 10 Hz on an onboard embedded computer.","It handles reactive avoidance of dynamic obstacles using both occupancy grids and LiDAR sensing.","The approach works for autonomous surface vehicles in simulation and non-holonomic mobile robots on hardware."],"fun_headline_variants":["Discrete-time control barrier functions for polytope containment","Polytope containment formulated as CBF constraints in MPC","Free-space polytope containment via barrier functions in controller","Barrier function constraints for containment without obstacle detection"],"cache_read_input_tokens":2112,"weakest_assumption_plain":"The continuously updated free-space region can be represented as a convex polytope containing the robot footprint.","fun_headline_variants_meta":{"raw":{"variants":["Discrete-time control barrier functions for polytope containment","Polytope containment formulated as CBF constraints in MPC","Free-space polytope containment via barrier functions in controller","Barrier function constraints for containment without obstacle detection"]},"model":"grok-4.3","cost_usd":0.006954,"raw_usage":{"total_tokens":3130,"prompt_tokens":643,"num_sources_used":0,"completion_tokens":57,"cost_in_usd_ticks":69540500,"prompt_tokens_details":{"text_tokens":643,"audio_tokens":0,"image_tokens":0,"cached_tokens":64},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":2430,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":643,"tokens_out":57,"duration_ms":16402,"temperature":1.0,"reasoning_tokens":2430,"cache_read_input_tokens":64,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-06-27T16:33:53.346259+00:00","model_set":{"reader":"grok-4.3"},"falsifier":"A scenario with many obstacles where the computation time does not decrease substantially compared to obstacle-based methods or where the robot violates containment despite the constraints being active.","supporting_citations":[],"review_version":1}