Skip to content
#small language model Open access

The Number of Directions You Can Push Is Not the Dimension of Where You Can Go ── going forward, turning, backing and turning back by epsilon each moves it in the unpushed sideways direction, with sideways displacement over epsilon^2 at -0.998334 for epsilon=0.1 and -1.000000 for 0.001 ── the separator is whether the discrepancy from alternating two pushes points outside the directions that can be pushed ── [Paper 777]

Sep 2026 · Zenodo (CERN European Organization for Nuclear Research)

Abstract

A car cannot move sideways. It can only drive forward or back and turn. Yet it can pull into a parking space beside it. How a third direction becomes reachable with only two directions of push is counted. No new theorem or law is claimed. Scope of this paper (scope note): No new theorem or law is claimed──the kinematic car model, non-holonomic constraints, the Lie bracket, reachability when brackets span all dimensions (Chow's theorem), and sideways motion by small back-and-forth manoeuvres are all known (Murray et al. 1994, LaValle 2006). Kinematics only──mass, inertia, slip and tyre forces are not treated; only a velocity-driven model is used. No theorem is proved──Chow's theorem and higher brackets are only named. No procedure is recommended──real parallel-parking procedures are not treated. Checked numerically──the loop was integrated by the midpoint method, and the bracket found by finite differences. Relation to earlier papers: Paper 601 showed that half of the fifteen puzzle's positions can never be reached──there an invariant attached to the moves put half the positions out of reach. Here, with only two directions of push, the discrepancy of alternating pushes makes a third direction and everything is reached. An invariant that makes an unreachable side, and a bracket that enlarges the reachable side. What is added is integrating the loop's displacement at three values of epsilon and showing it proportional to epsilon^2, checking the numerical Lie bracket against the formula and deriving a determinant of 1.000000, comparing with a cart whose bracket is 0, and placing the separator on whether the bracket points outside the pushable directions. First, two directions to push, three dimensions to reach──three states, position and heading; two inputs, forward and turning (Section 2). Second, alternating pushes moves the car in an unpushed direction──over one loop of forward, turn, back, turn back, sideways displacement over epsilon^2 goes from -0.998334 to -1.000000 (Section 2). Third, the Lie bracket sets that direction──the numerical bracket (0.644218, -0.764842, 0) matches the formula to 2.184x10^-11 (Section 3). Fourth, and this is the core. If the bracket points outside the pushable directions, every direction is reachable──the determinant across three directions is 1.000000 (Section 3). Fifth, if the bracket is 0, some direction stays out of reach──a cart pushed only in x and y has bracket 0 and cannot reach other headings (Section 3). Sixth, the separator is whether the discrepancy from alternating two pushes (the Lie bracket) points outside the directions that can be pushed──"only two controls, so only two dimensions of motion" has no truth value until the bracket is stated (Section 4). The number of directions you can push is not the dimension of where you can go. A car model has three states, position and heading, but only two inputs, forward motion and turning──yet one loop of forward, turn, back and turn back moves it sideways, a direction never pushed, with sideways displacement over epsilon^2 approaching -1.000000 at epsilon=0.001 from -0.998334 at 0.1. That direction is the Lie bracket, whose numerical value (0.644218, -0.764842, 0) matches the formula to 2.184x10^-11, and with g_1 and g_2 gives a determinant of 1.000000, spanning all three directions; a cart that cannot turn has bracket 0 and cannot reach other headings──the reachable dimension is set not by the number of pushes but by whether they fail to commute. The separator is whether the discrepancy from alternating two pushes (the Lie bracket) points outside the directions that can be pushed. Placed among the earlier papers──Paper 601 showed an invariant cutting the reachable positions in half; this counts a bracket enlarging the reachable directions. To be honest──only a kinematic model is used, with no mass, slip, proofs or real parking. On the making of this work: The ideas and content of this work stem from the author's own considerations. Assistance from an AI (a large language model) was used for structuring, English translation, and checking the algebra. Any remaining errors or misinterpretations are solely the author's. Feedback and corrections are sincerely appreciated. Keywords: underactuated systems, non-holonomic constraints, Lie bracket, reachability, car kinematics, parallel parking. ----- 車は横には動けない。動かせるのは前後と、ハンドルによる回転の二つだけである。それでも車は横に並んだ駐車場所に入れる。押せる向きが二つしかないのに、どうして三つ目の向きに行けるかを数える。新しい定理も法則も主張しない。 本稿の射程(射程注記):新しい定理も法則も主張しない──車の運動学の模型、非ホロノミックな拘束、リー括弧、括弧を足した向きが全次元に広がれば到達できること(チャウの定理)、小刻みの往復で横に動けることは、いずれも既知である(マレーほか 1994、ラヴァル 2006)。運動学だけを見る──質量、慣性、滑り、タイヤの力は扱わず、速度で動く模型だけを使う。定理を証明しない──チャウの定理や高次の括弧は名前を挙げるにとどめる。手順を勧めない──実在の車の縦列駐車の手順は扱わない。数値で確かめる──一回りの動きは中点法で積分し、括弧は差分で求めた。既刊との関係:論文601 は、15 パズルの局面の半分が、どう動かしても届かないことを示した──そこでは動かし方に付いてくる不変量が、局面の半分を届かなくした。ここでは押せる向きが二つしかないのに、交互の押し方の食い違いが三つ目の向きを作り、全部に届く。届かない側を作る不変量と、届く側を増やす括弧になる。加えたのは、一回りの動きのずれを epsilon の三つの値で積分し、epsilon^2 に比例することを示したこと、数値のリー括弧を式と照らし、行列式 1.000000 を出したこと、括弧が 0 の台車と比べたこと、分離子を「括弧が押せる向きの外に出るか」に置いたことである。 第一に、押せる向きは二つ、行きたい場所は三次元──状態は位置と向きの三つ、入力は前後と回転の二つ(第2節)。 第二に、交互に押すと、押していない向きに動く──前・回る・後ろ・戻るの一回りで、横のずれ/epsilon^2 は -0.998334 から -1.000000 へ(第2節)。 第三に、その向きは、リー括弧が決める──数値の括弧 (0.644218, -0.764842, 0) が式と差 2.184x10^-11 で一致(第3節)。 第四に、これが本稿の芯である。括弧が押せる向きの外に出れば、全部の向きに行ける──三方向の行列式は 1.000000(第3節)。 第五に、括弧が 0 なら、行けない向きが残る──x と y にだけ押せる台車は括弧が 0 で、向き theta には行けない(第3節)。 第六に、分離子は、二つの押し方を交互に繰り返したときの食い違い(リー括弧)が、押せる向きの外に出るかである──「二つしか動かせないから、二次元しか動けない」は、括弧を言うまで真偽が決まらない(第4節)。 押せる向きの数と、行ける場所の次元は違う。車の模型は位置と向きの三つの状態に対し、動かせるのは前後と回転の二つだけである──それでも前・回る・後ろ・戻るの一回りで、押していない横向きへ動き、横のずれ/epsilon^2 は epsilon=0.1 の -0.998334 から 0.001 の -1.000000 へ近づく。その向きはリー括弧で、数値 (0.644218, -0.764842, 0) が式と差 2.184x10^-11 で一致し、g_1・g_2 と合わせた行列式は 1.000000 で三つの向きがそろう。向きを変えられない台車では括弧が 0 で、向きには行けない──行ける次元を決めているのは、押せる向きの数ではなく、押し方の食い違いである。分離子は、二つの押し方を交互に繰り返したときの食い違い(リー括弧)が、押せる向きの外に出るかである。既刊との位置──論文601 は不変量が届く局面を半分に削ることを示した。ここは括弧が届く向きを増やすことを数えた。正直に言えば──運動学の模型だけで、質量も滑りも定理の証明も実在の駐車も扱っていない。 作成にあたって:本稿の着想と内容は、著者自身の考察に基づくものです。文章の構成整理や英訳、数式の確認には AI(大規模言語モデル)の助力を得ました。最終的な内容の解釈や誤りがあれば、それらはすべて著者の責に帰します。お気づきの点があれば、ご教示いただければ幸いです。 キーワード:劣駆動系、非ホロノミック拘束、リー括弧、到達可能性、車の運動学、縦列駐車。

View source

Similar papers

#small language model Dataset Open access Oct 2026

Socratic guiding questions in synthetic arithmetic data: matched LoRA runs (revision v2)

Supporting data, adapters, predictions and code for the article *Low-Cost LoRA Fine-Tuning of Small Language Models for Multi-Step Arithmetic Reasoning* by Jake O'Grady, Asena Isik Gürhan, Chee Fong Ting and Effirul Ramlan (University of Galway). We generated 20,000 GSM8K-derived arithmetic problems with step-by-step s...

O'Grady, Jake, Gürhan, Asena Isik, Chee, Fong Ting et al. · 465 citations
#computer vision Open access Jun 2016

Software Development in Startup Companies: The Greenfield Startup Model

The results are packaged in the Greenfield Startup Model (GSM), which explains the priority of startups to release the product as quickly as possible, and the need to shorten time-to-market, by speeding up the development through low-precision engineering activities.

Carmine Giardino, Nicolò Paternoster, M. Unterkalmsteiner et al. · 178 citations · ⚡14
#computer vision Open access Oct 2016

Software Startups - A Research Agenda

Software startup companies develop innovative, software-intensive products within limited timeframes and with few resources, searching for sustainable and scalable business models.

M. Unterkalmsteiner, P. Abrahamsson, Xiaofeng Wang et al. · 157 citations · ⚡17
#machine learning Review Open access Oct 2016

“Failures” to be celebrated: an analysis of major pivots of software startups

This study conducts a case survey study based on the secondary data of the major pivots happened in 49 software startups, and demonstrates that customer need pivot is the most common among all pivot types.

Sohaib Shahid Bajwa, Xiaofeng Wang, Anh Nguyen-Duc et al. · 127 citations · ⚡15
#computer vision Review Open access May 2015

A survey study on major technical barriers affecting the decision to adopt cloud services

The comparison of adopter and non-adopter sample reveals three potential adoption inhibitor, security, data privacy, and portability, which underlines the importance of the technical and security perspectives for research investigating the adoption of technology.

Nattakarn Phaphoom, Xiaofeng Wang, S. Samuel et al. · 111 citations · ⚡8
#computer vision Open access Feb 2018

Lean Internal Startups for Software Product Innovation in Large Companies: Enablers and Inhibitors

This study investigates how Lean internal startup facilitates software product innovation in large companies and identifies its enablers and inhibitors, and shows the potential of the method-in-action framework to investigate the Lean startup approach in non-startup context.

Henry Edison, Nina M. Smørsgård, Xiaofeng Wang et al. · 78 citations · ⚡6

Related blog posts

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.